MCT USE ONLY.

STUDENT USE PROHIBITED

O F F I C I A L M I C R O S O F T L E A R N I N G P R O D U C T

20768C
Developing SQL Data Models
 MCT USE ONLY. STUDENT USE PROHIBITED
ii Developing SQL Data Models

Information in this document, including URL and other Internet Web site references, is subject to change
without notice. Unless otherwise noted, the example companies, organizations, products, domain names,
e-mail addresses, logos, people, places, and events depicted herein are fictitious, and no association with
any real company, organization, product, domain name, e-mail address, logo, person, place or event is
intended or should be inferred. Complying with all applicable copyright laws is the responsibility of the
user. Without limiting the rights under copyright, no part of this document may be reproduced, stored in
or introduced into a retrieval system, or transmitted in any form or by any means (electronic, mechanical,
photocopying, recording, or otherwise), or for any purpose, without the express written permission of
Microsoft Corporation.

Microsoft may have patents, patent applications, trademarks, copyrights, or other intellectual property
rights covering subject matter in this document. Except as expressly provided in any written license
agreement from Microsoft, the furnishing of this document does not give you any license to these
patents, trademarks, copyrights, or other intellectual property.

The names of manufacturers, products, or URLs are provided for informational purposes only and
Microsoft makes no representations and warranties, either expressed, implied, or statutory, regarding
these manufacturers or the use of the products with any Microsoft technologies. The inclusion of a
manufacturer or product does not imply endorsement of Microsoft of the manufacturer or product. Links
may be provided to third party sites. Such sites are not under the control of Microsoft and Microsoft is not
responsible for the contents of any linked site or any link contained in a linked site, or any changes or
updates to such sites. Microsoft is not responsible for webcasting or any other form of transmission
received from any linked site. Microsoft is providing these links to you only as a convenience, and the
inclusion of any link does not imply endorsement of Microsoft of the site or the products contained
therein.
© 2018 Microsoft Corporation. All rights reserved.

Microsoft and the trademarks listed at

https://www.microsoft.com/en-us/legal/intellectualproperty/trademarks/en-us.aspx are trademarks of the
Microsoft group of companies. All other trademarks are property of their respective owners.

Product Number: 20768C

Part Number (if applicable): X21-64459

Released: 02/2018
 MCT USE ONLY. STUDENT USE PROHIBITED
MICROSOFT LICENSE TERMS
MICROSOFT INSTRUCTOR-LED COURSEWARE

These license terms are an agreement between Microsoft Corporation (or based on where you live, one of its
affiliates) and you. Please read them. They apply to your use of the content accompanying this agreement which
includes the media on which you received it, if any. These license terms also apply to Trainer Content and any
updates and supplements for the Licensed Content unless other terms accompany those items. If so, those terms
apply.

BY ACCESSING, DOWNLOADING OR USING THE LICENSED CONTENT, YOU ACCEPT THESE TERMS.
IF YOU DO NOT ACCEPT THEM, DO NOT ACCESS, DOWNLOAD OR USE THE LICENSED CONTENT.

If you comply with these license terms, you have the rights below for each license you acquire.

1. DEFINITIONS.

a. “Authorized Learning Center” means a Microsoft IT Academy Program Member, Microsoft Learning
Competency Member, or such other entity as Microsoft may designate from time to time.

b. “Authorized Training Session” means the instructor-led training class using Microsoft Instructor-Led
Courseware conducted by a Trainer at or through an Authorized Learning Center.

c. “Classroom Device” means one (1) dedicated, secure computer that an Authorized Learning Center owns
or controls that is located at an Authorized Learning Center’s training facilities that meets or exceeds the
hardware level specified for the particular Microsoft Instructor-Led Courseware.

d. “End User” means an individual who is (i) duly enrolled in and attending an Authorized Training Session
or Private Training Session, (ii) an employee of a MPN Member, or (iii) a Microsoft full-time employee.

e. “Licensed Content” means the content accompanying this agreement which may include the Microsoft
Instructor-Led Courseware or Trainer Content.

f. “Microsoft Certified Trainer” or “MCT” means an individual who is (i) engaged to teach a training session
to End Users on behalf of an Authorized Learning Center or MPN Member, and (ii) currently certified as a
Microsoft Certified Trainer under the Microsoft Certification Program.

g. “Microsoft Instructor-Led Courseware” means the Microsoft-branded instructor-led training course that
educates IT professionals and developers on Microsoft technologies. A Microsoft Instructor-Led
Courseware title may be branded as MOC, Microsoft Dynamics or Microsoft Business Group courseware.

h. “Microsoft IT Academy Program Member” means an active member of the Microsoft IT Academy
Program.

i. “Microsoft Learning Competency Member” means an active member of the Microsoft Partner Network
program in good standing that currently holds the Learning Competency status.

j. “MOC” means the “Official Microsoft Learning Product” instructor-led courseware known as Microsoft
Official Course that educates IT professionals and developers on Microsoft technologies.

k. “MPN Member” means an active Microsoft Partner Network program member in good standing.
 MCT USE ONLY. STUDENT USE PROHIBITED
l. “Personal Device” means one (1) personal computer, device, workstation or other digital electronic device
that you personally own or control that meets or exceeds the hardware level specified for the particular
Microsoft Instructor-Led Courseware.

m. “Private Training Session” means the instructor-led training classes provided by MPN Members for
corporate customers to teach a predefined learning objective using Microsoft Instructor-Led Courseware.
These classes are not advertised or promoted to the general public and class attendance is restricted to
individuals employed by or contracted by the corporate customer.

n. “Trainer” means (i) an academically accredited educator engaged by a Microsoft IT Academy Program
Member to teach an Authorized Training Session, and/or (ii) a MCT.

o. “Trainer Content” means the trainer version of the Microsoft Instructor-Led Courseware and additional
supplemental content designated solely for Trainers’ use to teach a training session using the Microsoft
Instructor-Led Courseware. Trainer Content may include Microsoft PowerPoint presentations, trainer
preparation guide, train the trainer materials, Microsoft One Note packs, classroom setup guide and Pre-
release course feedback form. To clarify, Trainer Content does not include any software, virtual hard
disks or virtual machines.

2. USE RIGHTS. The Licensed Content is licensed not sold. The Licensed Content is licensed on a one copy
per user basis, such that you must acquire a license for each individual that accesses or uses the Licensed
Content.

2.1 Below are five separate sets of use rights. Only one set of rights apply to you.

a. If you are a Microsoft IT Academy Program Member:

i. Each license acquired on behalf of yourself may only be used to review one (1) copy of the Microsoft
Instructor-Led Courseware in the form provided to you. If the Microsoft Instructor-Led Courseware is
in digital format, you may install one (1) copy on up to three (3) Personal Devices. You may not
install the Microsoft Instructor-Led Courseware on a device you do not own or control.
ii. For each license you acquire on behalf of an End User or Trainer, you may either:
1. distribute one (1) hard copy version of the Microsoft Instructor-Led Courseware to one (1) End
User who is enrolled in the Authorized Training Session, and only immediately prior to the
commencement of the Authorized Training Session that is the subject matter of the Microsoft
Instructor-Led Courseware being provided, or
2. provide one (1) End User with the unique redemption code and instructions on how they can
access one (1) digital version of the Microsoft Instructor-Led Courseware, or
3. provide one (1) Trainer with the unique redemption code and instructions on how they can
access one (1) Trainer Content,
provided you comply with the following:
iii. you will only provide access to the Licensed Content to those individuals who have acquired a valid
license to the Licensed Content,
iv. you will ensure each End User attending an Authorized Training Session has their own valid licensed
copy of the Microsoft Instructor-Led Courseware that is the subject of the Authorized Training
Session,
v. you will ensure that each End User provided with the hard-copy version of the Microsoft Instructor-
Led Courseware will be presented with a copy of this agreement and each End User will agree that
their use of the Microsoft Instructor-Led Courseware will be subject to the terms in this agreement
prior to providing them with the Microsoft Instructor-Led Courseware. Each individual will be required
to denote their acceptance of this agreement in a manner that is enforceable under local law prior to
their accessing the Microsoft Instructor-Led Courseware,
vi. you will ensure that each Trainer teaching an Authorized Training Session has their own valid
licensed copy of the Trainer Content that is the subject of the Authorized Training Session,
 MCT USE ONLY. STUDENT USE PROHIBITED
vii. you will only use qualified Trainers who have in-depth knowledge of and experience with the
Microsoft technology that is the subject of the Microsoft Instructor-Led Courseware being taught for
all your Authorized Training Sessions,
viii. you will only deliver a maximum of 15 hours of training per week for each Authorized Training
Session that uses a MOC title, and
ix. you acknowledge that Trainers that are not MCTs will not have access to all of the trainer resources
for the Microsoft Instructor-Led Courseware.

b. If you are a Microsoft Learning Competency Member:

i. Each license acquired on behalf of yourself may only be used to review one (1) copy of the Microsoft
Instructor-Led Courseware in the form provided to you. If the Microsoft Instructor-Led Courseware is
in digital format, you may install one (1) copy on up to three (3) Personal Devices. You may not
install the Microsoft Instructor-Led Courseware on a device you do not own or control.
ii. For each license you acquire on behalf of an End User or Trainer, you may either:
1. distribute one (1) hard copy version of the Microsoft Instructor-Led Courseware to one (1) End
User attending the Authorized Training Session and only immediately prior to the
commencement of the Authorized Training Session that is the subject matter of the Microsoft
Instructor-Led Courseware provided, or
2. provide one (1) End User attending the Authorized Training Session with the unique redemption
code and instructions on how they can access one (1) digital version of the Microsoft Instructor-
Led Courseware, or
3. you will provide one (1) Trainer with the unique redemption code and instructions on how they
can access one (1) Trainer Content,
provided you comply with the following:
iii. you will only provide access to the Licensed Content to those individuals who have acquired a valid
license to the Licensed Content,
iv. you will ensure that each End User attending an Authorized Training Session has their own valid
licensed copy of the Microsoft Instructor-Led Courseware that is the subject of the Authorized
Training Session,
v. you will ensure that each End User provided with a hard-copy version of the Microsoft Instructor-Led
Courseware will be presented with a copy of this agreement and each End User will agree that their
use of the Microsoft Instructor-Led Courseware will be subject to the terms in this agreement prior to
providing them with the Microsoft Instructor-Led Courseware. Each individual will be required to
denote their acceptance of this agreement in a manner that is enforceable under local law prior to
their accessing the Microsoft Instructor-Led Courseware,
vi. you will ensure that each Trainer teaching an Authorized Training Session has their own valid
licensed copy of the Trainer Content that is the subject of the Authorized Training Session,
vii. you will only use qualified Trainers who hold the applicable Microsoft Certification credential that is
the subject of the Microsoft Instructor-Led Courseware being taught for your Authorized Training
Sessions,
viii. you will only use qualified MCTs who also hold the applicable Microsoft Certification credential that is
the subject of the MOC title being taught for all your Authorized Training Sessions using MOC,
ix. you will only provide access to the Microsoft Instructor-Led Courseware to End Users, and
x. you will only provide access to the Trainer Content to Trainers.
 MCT USE ONLY. STUDENT USE PROHIBITED
c. If you are a MPN Member:
i. Each license acquired on behalf of yourself may only be used to review one (1) copy of the Microsoft
Instructor-Led Courseware in the form provided to you. If the Microsoft Instructor-Led Courseware is
in digital format, you may install one (1) copy on up to three (3) Personal Devices. You may not
install the Microsoft Instructor-Led Courseware on a device you do not own or control.
ii. For each license you acquire on behalf of an End User or Trainer, you may either:
1. distribute one (1) hard copy version of the Microsoft Instructor-Led Courseware to one (1) End
User attending the Private Training Session, and only immediately prior to the commencement
of the Private Training Session that is the subject matter of the Microsoft Instructor-Led
Courseware being provided, or
2. provide one (1) End User who is attending the Private Training Session with the unique
redemption code and instructions on how they can access one (1) digital version of the
Microsoft Instructor-Led Courseware, or
3. you will provide one (1) Trainer who is teaching the Private Training Session with the unique
redemption code and instructions on how they can access one (1) Trainer Content,
provided you comply with the following:
iii. you will only provide access to the Licensed Content to those individuals who have acquired a valid
license to the Licensed Content,
iv. you will ensure that each End User attending an Private Training Session has their own valid licensed
copy of the Microsoft Instructor-Led Courseware that is the subject of the Private Training Session,
v. you will ensure that each End User provided with a hard copy version of the Microsoft Instructor-Led
Courseware will be presented with a copy of this agreement and each End User will agree that their
use of the Microsoft Instructor-Led Courseware will be subject to the terms in this agreement prior to
providing them with the Microsoft Instructor-Led Courseware. Each individual will be required to
denote their acceptance of this agreement in a manner that is enforceable under local law prior to
their accessing the Microsoft Instructor-Led Courseware,
vi. you will ensure that each Trainer teaching an Private Training Session has their own valid licensed
copy of the Trainer Content that is the subject of the Private Training Session,
vii. you will only use qualified Trainers who hold the applicable Microsoft Certification credential that is
the subject of the Microsoft Instructor-Led Courseware being taught for all your Private Training
Sessions,
viii. you will only use qualified MCTs who hold the applicable Microsoft Certification credential that is the
subject of the MOC title being taught for all your Private Training Sessions using MOC,
ix. you will only provide access to the Microsoft Instructor-Led Courseware to End Users, and
x. you will only provide access to the Trainer Content to Trainers.

d. If you are an End User:

For each license you acquire, you may use the Microsoft Instructor-Led Courseware solely for your
personal training use. If the Microsoft Instructor-Led Courseware is in digital format, you may access the
Microsoft Instructor-Led Courseware online using the unique redemption code provided to you by the
training provider and install and use one (1) copy of the Microsoft Instructor-Led Courseware on up to
three (3) Personal Devices. You may also print one (1) copy of the Microsoft Instructor-Led Courseware.
You may not install the Microsoft Instructor-Led Courseware on a device you do not own or control.

e. If you are a Trainer.

i. For each license you acquire, you may install and use one (1) copy of the Trainer Content in the
form provided to you on one (1) Personal Device solely to prepare and deliver an Authorized
Training Session or Private Training Session, and install one (1) additional copy on another Personal
Device as a backup copy, which may be used only to reinstall the Trainer Content. You may not
install or use a copy of the Trainer Content on a device you do not own or control. You may also
print one (1) copy of the Trainer Content solely to prepare for and deliver an Authorized Training
Session or Private Training Session.
 MCT USE ONLY. STUDENT USE PROHIBITED
ii. You may customize the written portions of the Trainer Content that are logically associated with
instruction of a training session in accordance with the most recent version of the MCT agreement.
If you elect to exercise the foregoing rights, you agree to comply with the following: (i)
customizations may only be used for teaching Authorized Training Sessions and Private Training
Sessions, and (ii) all customizations will comply with this agreement. For clarity, any use of
“customize” refers only to changing the order of slides and content, and/or not using all the slides or
content, it does not mean changing or modifying any slide or content.

2.2 Separation of Components. The Licensed Content is licensed as a single unit and you may not
separate their components and install them on different devices.

2.3 Redistribution of Licensed Content. Except as expressly provided in the use rights above, you may
not distribute any Licensed Content or any portion thereof (including any permitted modifications) to any
third parties without the express written permission of Microsoft.

2.4 Third Party Notices. The Licensed Content may include third party code tent that Microsoft, not the
third party, licenses to you under this agreement. Notices, if any, for the third party code ntent are included
for your information only.

2.5 Additional Terms. Some Licensed Content may contain components with additional terms,
conditions, and licenses regarding its use. Any non-conflicting terms in those conditions and licenses also
apply to your use of that respective component and supplements the terms described in this agreement.

3. LICENSED CONTENT BASED ON PRE-RELEASE TECHNOLOGY. If the Licensed Content’s subject

matter is based on a pre-release version of Microsoft technology (“Pre-release”), then in addition to the
other provisions in this agreement, these terms also apply:

a. Pre-Release Licensed Content. This Licensed Content subject matter is on the Pre-release version of
the Microsoft technology. The technology may not work the way a final version of the technology will
and we may change the technology for the final version. We also may not release a final version.
Licensed Content based on the final version of the technology may not contain the same information as
the Licensed Content based on the Pre-release version. Microsoft is under no obligation to provide you
with any further content, including any Licensed Content based on the final version of the technology.

b. Feedback. If you agree to give feedback about the Licensed Content to Microsoft, either directly or
through its third party designee, you give to Microsoft without charge, the right to use, share and
commercialize your feedback in any way and for any purpose. You also give to third parties, without
charge, any patent rights needed for their products, technologies and services to use or interface with
any specific parts of a Microsoft technology, Microsoft product, or service that includes the feedback.
You will not give feedback that is subject to a license that requires Microsoft to license its technology,
technologies, or products to third parties because we include your feedback in them. These rights
survive this agreement.

c. Pre-release Term. If you are an Microsoft IT Academy Program Member, Microsoft Learning
Competency Member, MPN Member or Trainer, you will cease using all copies of the Licensed Content on
the Pre-release technology upon (i) the date which Microsoft informs you is the end date for using the
Licensed Content on the Pre-release technology, or (ii) sixty (60) days after the commercial release of the
technology that is the subject of the Licensed Content, whichever is earliest (“Pre-release term”).
Upon expiration or termination of the Pre-release term, you will irretrievably delete and destroy all copies
of the Licensed Content in your possession or under your control.
 MCT USE ONLY. STUDENT USE PROHIBITED
4. SCOPE OF LICENSE. The Licensed Content is licensed, not sold. This agreement only gives you some
rights to use the Licensed Content. Microsoft reserves all other rights. Unless applicable law gives you more
rights despite this limitation, you may use the Licensed Content only as expressly permitted in this
agreement. In doing so, you must comply with any technical limitations in the Licensed Content that only
allows you to use it in certain ways. Except as expressly permitted in this agreement, you may not:
• access or allow any individual to access the Licensed Content if they have not acquired a valid license
for the Licensed Content,
• alter, remove or obscure any copyright or other protective notices (including watermarks), branding
or identifications contained in the Licensed Content,
• modify or create a derivative work of any Licensed Content,
• publicly display, or make the Licensed Content available for others to access or use,
• copy, print, install, sell, publish, transmit, lend, adapt, reuse, link to or post, make available or
distribute the Licensed Content to any third party,
• work around any technical limitations in the Licensed Content, or
• reverse engineer, decompile, remove or otherwise thwart any protections or disassemble the
Licensed Content except and only to the extent that applicable law expressly permits, despite this
limitation.

5. RESERVATION OF RIGHTS AND OWNERSHIP. Microsoft reserves all rights not expressly granted to
you in this agreement. The Licensed Content is protected by copyright and other intellectual property laws
and treaties. Microsoft or its suppliers own the title, copyright, and other intellectual property rights in the
Licensed Content.

6. EXPORT RESTRICTIONS. The Licensed Content is subject to United States export laws and regulations.
You must comply with all domestic and international export laws and regulations that apply to the Licensed
Content. These laws include restrictions on destinations, end users and end use. For additional information,
see www.microsoft.com/exporting.

7. SUPPORT SERVICES. Because the Licensed Content is “as is”, we may not provide support services for it.

8. TERMINATION. Without prejudice to any other rights, Microsoft may terminate this agreement if you fail
to comply with the terms and conditions of this agreement. Upon termination of this agreement for any
reason, you will immediately stop all use of and delete and destroy all copies of the Licensed Content in
your possession or under your control.

9. LINKS TO THIRD PARTY SITES. You may link to third party sites through the use of the Licensed
Content. The third party sites are not under the control of Microsoft, and Microsoft is not responsible for
the contents of any third party sites, any links contained in third party sites, or any changes or updates to
third party sites. Microsoft is not responsible for webcasting or any other form of transmission received
from any third party sites. Microsoft is providing these links to third party sites to you only as a
convenience, and the inclusion of any link does not imply an endorsement by Microsoft of the third party
site.

10. ENTIRE AGREEMENT. This agreement, and any additional terms for the Trainer Content, updates and
supplements are the entire agreement for the Licensed Content, updates and supplements.

11. APPLICABLE LAW.

a. United States. If you acquired the Licensed Content in the United States, Washington state law governs
the interpretation of this agreement and applies to claims for breach of it, regardless of conflict of laws
principles. The laws of the state where you live govern all other claims, including claims under state
consumer protection laws, unfair competition laws, and in tort.
 MCT USE ONLY. STUDENT USE PROHIBITED
b. Outside the United States. If you acquired the Licensed Content in any other country, the laws of that
country apply.

12. LEGAL EFFECT. This agreement describes certain legal rights. You may have other rights under the laws
of your country. You may also have rights with respect to the party from whom you acquired the Licensed
Content. This agreement does not change your rights under the laws of your country if the laws of your
country do not permit it to do so.

13. DISCLAIMER OF WARRANTY. THE LICENSED CONTENT IS LICENSED "AS-IS" AND "AS
AVAILABLE." YOU BEAR THE RISK OF USING IT. MICROSOFT AND ITS RESPECTIVE
AFFILIATES GIVES NO EXPRESS WARRANTIES, GUARANTEES, OR CONDITIONS. YOU MAY
HAVE ADDITIONAL CONSUMER RIGHTS UNDER YOUR LOCAL LAWS WHICH THIS AGREEMENT
CANNOT CHANGE. TO THE EXTENT PERMITTED UNDER YOUR LOCAL LAWS, MICROSOFT AND
ITS RESPECTIVE AFFILIATES EXCLUDES ANY IMPLIED WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT.

14. LIMITATION ON AND EXCLUSION OF REMEDIES AND DAMAGES. YOU CAN RECOVER FROM
MICROSOFT, ITS RESPECTIVE AFFILIATES AND ITS SUPPLIERS ONLY DIRECT DAMAGES UP
TO US$5.00. YOU CANNOT RECOVER ANY OTHER DAMAGES, INCLUDING CONSEQUENTIAL,
LOST PROFITS, SPECIAL, INDIRECT OR INCIDENTAL DAMAGES.

This limitation applies to

o anything related to the Licensed Content, services, content (including code) on third party Internet
sites or third-party programs; and
o claims for breach of contract, breach of warranty, guarantee or condition, strict liability, negligence,
or other tort to the extent permitted by applicable law.

It also applies even if Microsoft knew or should have known about the possibility of the damages. The
above limitation or exclusion may not apply to you because your country may not allow the exclusion or
limitation of incidental, consequential or other damages.

Please note: As this Licensed Content is distributed in Quebec, Canada, some of the clauses in this
agreement are provided below in French.

Remarque : Ce le contenu sous licence étant distribué au Québec, Canada, certaines des clauses
dans ce contrat sont fournies ci-dessous en français.

EXONÉRATION DE GARANTIE. Le contenu sous licence visé par une licence est offert « tel quel ». Toute
utilisation de ce contenu sous licence est à votre seule risque et péril. Microsoft n’accorde aucune autre garantie
expresse. Vous pouvez bénéficier de droits additionnels en vertu du droit local sur la protection dues
consommateurs, que ce contrat ne peut modifier. La ou elles sont permises par le droit locale, les garanties
implicites de qualité marchande, d’adéquation à un usage particulier et d’absence de contrefaçon sont exclues.

LIMITATION DES DOMMAGES-INTÉRÊTS ET EXCLUSION DE RESPONSABILITÉ POUR LES

DOMMAGES. Vous pouvez obtenir de Microsoft et de ses fournisseurs une indemnisation en cas de dommages
directs uniquement à hauteur de 5,00 $ US. Vous ne pouvez prétendre à aucune indemnisation pour les autres
dommages, y compris les dommages spéciaux, indirects ou accessoires et pertes de bénéfices.
Cette limitation concerne:
• tout ce qui est relié au le contenu sous licence, aux services ou au contenu (y compris le code)
figurant sur des sites Internet tiers ou dans des programmes tiers; et.
• les réclamations au titre de violation de contrat ou de garantie, ou au titre de responsabilité
stricte, de négligence ou d’une autre faute dans la limite autorisée par la loi en vigueur.
 MCT USE ONLY. STUDENT USE PROHIBITED
Elle s’applique également, même si Microsoft connaissait ou devrait connaître l’éventualité d’un tel dommage. Si
votre pays n’autorise pas l’exclusion ou la limitation de responsabilité pour les dommages indirects, accessoires
ou de quelque nature que ce soit, il se peut que la limitation ou l’exclusion ci-dessus ne s’appliquera pas à votre
égard.

EFFET JURIDIQUE. Le présent contrat décrit certains droits juridiques. Vous pourriez avoir d’autres droits
prévus par les lois de votre pays. Le présent contrat ne modifie pas les droits que vous confèrent les lois de votre
pays si celles-ci ne le permettent pas.

Revised July 2013

MCT USE ONLY. STUDENT USE PROHIBITED
xi
Developing SQL Data Models
 MCT USE ONLY. STUDENT USE PROHIBITED
xii Developing SQL Data Models

Acknowledgements
Microsoft Learning would like to acknowledge and thank the following for their contribution towards
developing this title. Their effort at various stages in the development has ensured that you have a good
classroom experience.

Alistair Matthews – Content Developer

Alistair Matthews is an experienced consultant and technical author who has worked with a wide variety
of Microsoft technologies over his nineteen-year career. Recent projects have included revising SQL Server
2014 courseware, authorizing Azure courseware, and developing SharePoint apps, all for different groups
within Microsoft. He designed and wrote the Microsoft Official Curriculum course number 20486
"Developing ASP.NET 4.5 MVC Web Applications" and several modules in the SharePoint developers'
courses numbers 20488 and 20489. He has developed content for Microsoft Press books, TechNet, and
MSDN and has written e-learning courseware and examination questions. He lives in Cornwall in the
United Kingdom.

Ed Harper – Content Developer

Ed Harper is a database developer specializing in Microsoft SQL Server. Ed has worked with SQL Server
since 1999, and has developed and designed transaction-processing and reporting systems for cloud
security, telecommunications, and financial services companies.

Jamie Newman – Content Developer

Jamie Newman became an IT trainer in 1997, first for an IT training company and later for a university,
where he became involved in developing courses as well as training them. He began to specialize in
databases and eventually moved into database consultancy. In recent years he has specialized in SQL
Server and has set up multi user systems that are accessed nationwide. Despite now being more involved
with development work, Jamie still likes to deliver IT training courses when the opportunity arises!

John Daisley – Content Developer

John Daisley is a mixed vendor Business Intelligence and Data Warehousing contractor with a wealth of
data warehousing and Microsoft SQL Server database administration experience. Having worked in the
Business Intelligence arena for over a decade, John has extensive experience of implementing business
intelligence and database solutions using a wide range of technologies and across a wide range of
industries including airlines, engineering, financial services, and manufacturing.

Martin Ellis – Content Developer

Martin is a technical author, consultant, and SQL Server subject matter expert. As a Microsoft Certified
Trainer (MCT), he has extensive experience of delivering Microsoft training courses on technologies
including SQL Server and Windows Server. He has authored and co-authored numerous Microsoft courses
for the SQL Server 2008 R2, SQL Server 2012, SQL Server 2014, and SQL Server 2016 curriculums, and is
the author of several white papers and other technical documents. Martin also provides consultancy
services for organizations in the planning, implementation, optimization, and maintenance of SQL Server
solutions.

Simon Butler – Content Developer

Simon Butler FISTC is a highly-experienced Senior Technical Writer with nearly 30 years' experience in the
profession. He has written training materials and other information products for several high-profile
clients. He is a Fellow of the Institute of Scientific and Technical Communicators (ISTC), the UK
professional body for Technical Writers/Authors. To gain this, his skills, experience and knowledge have
been judged and assessed by the Membership Panel. He is also a Past President of the Institute and has
been a tutor on the ISTC Open Learning course in Technical Communication techniques. His writing skills
are augmented by extensive technical skills gained within the computing and electronics fields.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models xiii

Geoff Allix – Technical Reviewer

Geoff Allix is a Microsoft SQL Server subject matter expert and professional content developer at Content
Master—a division of CM Group Ltd. As a Microsoft Certified Trainer, Geoff has delivered training courses
on SQL Server since version 6.5. Geoff is a Microsoft Certified IT Professional for SQL Server and has
extensive experience in designing and implementing database and BI solutions on SQL Server
technologies, and has provided consultancy services to organizations seeking to implement and optimize
database solutions.

Lin Joyner – Technical Reviewer

Lin is an experienced Microsoft SQL Server developer and administrator. She has worked with SQL Server
since version 6.0 and previously as a Microsoft Certified Trainer, delivered training courses across the UK.
Lin has a wide breadth of knowledge across SQL Server technologies, including BI and Reporting Services.
Lin also designs and authors SQL Server and .NET development training materials. She has been writing
instructional content for Microsoft for over 15 years.
 MCT USE ONLY. STUDENT USE PROHIBITED
xiv Developing SQL Data Models

Contents
Module 1: Introduction to Business Intelligence and Data Modeling
Module Overview 1-1 
Lesson 1: Overview of Business Intelligence and Data Models 1-2 

Lesson 2: Microsoft Business Intelligence Platform 1-7 

Lab: Exploring a Business Intelligence Solution 1-14 

Module Review and Takeaways 1-18 

Module 2: Creating Multidimensional Databases

Module Overview 2-1 

Lesson 1: Introduction to Multidimensional Analysis 2-2 

Lesson 2: Data Sources and Data Source Views 2-9 

Lesson 3: Cubes 2-13 

Lesson 4: Overview of Cube Security 2-19 

Lesson 5: Configuring SSAS 2-25 

Lesson 6: Monitoring SSAS 2-30 

Lab: Creating a Multidimensional Database 2-35 

Module Review and Takeaways 2-41 

Module 3: Working with Cubes and Dimensions

Module Overview 3-1 

Lesson 1: Configuring Dimensions 3-2 

Lesson 2: Defining Attribute Hierarchies 3-7 

Lesson 3: Implementing Sorting and Grouping for Attributes 3-13 

Lesson 4: Slowly Changing Dimensions 3-16 

Lab: Working with Cubes and Dimensions 3-19 

Module Review and Takeaways 3-24 

Module 4: Working with Measures and Measure Groups

Module Overview 4-1 

Lesson 1: Working with Measures 4-2 

Lesson 2: Working with Measure Groups 4-6 

Lab: Configuring Measures and Measure Groups 4-15 

Module Review and Takeaways 4-19 

 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models xv

Module 5: Introduction to MDX

Module Overview 5-1 

Lesson 1: MDX Fundamentals 5-2 

Lesson 2: Adding Calculations to a Cube 5-10 

Lesson 3: Using MDX to Query a Cube 5-17 

Lab: Using MDX 5-24 

Module Review and Takeaways 5-28 

Module 6: Customizing Cube Functionality

Module Overview 6-1 

Lesson 1: Implementing Key Performance Indicators 6-2 

Lesson 2: Implementing Actions 6-8 

Lesson 3: Implementing Perspectives 6-11 

Lesson 4: Implementing Translations 6-14 

Lab: Customizing a Cube 6-16 

Module Review and Takeaways 6-20 

Module 7: Implementing a Tabular Data Model by Using SQL Server Analysis Services
Module Overview 7-1 

Lesson 1: Introduction to Tabular Data Models in SQL Server Analysis Services 7-2 

Lesson 2: Creating a Tabular Data Model 7-7 

Lesson 3: Using a SQL Server Analysis Services Tabular Model in an Enterprise
BI Solution 7-18 

Lab: Implementing a Tabular Data Model in SQL Server Analysis Services 7-26 

Module Review and Takeaways 7-36 

Module 8: Introduction to Data Analysis Expressions (DAX)

Module Overview 8-1 

Lesson 1: DAX Fundamentals 8-2 

Lesson 2: Using DAX to Create Calculated Columns and Measures in a Tabular

Data Model 8-12 

Lab: Using DAX to Enhance a Tabular Data Model 8-24 

Module Review and Takeaways 8-30 

 MCT USE ONLY. STUDENT USE PROHIBITED
xvi Developing SQL Data Models

Module 9: Performing Predictive Analysis with Data Mining

Module Overview 9-1 

Lesson 1: Overview of Data Mining 9-2 

Lesson 2: Creating a Custom Data Mining Solution 9-7 

Lesson 3: Validating a Data Mining Model 9-15 

Lesson 4: Connecting to and Consuming a Data Mining Model 9-20 

Lab: Using Data Mining to Support a Marketing Campaign 9-23 

Module Review and Takeaways 9-28 

Lab Answer Keys

Module 1 Lab: Exploring a Business Intelligence Solution L01-1

Module 2 Lab: Creating a Multidimensional Database L02-1

Module 3 Lab: Working with Cubes and Dimensions L03-1

Module 4 Lab: Configuring Measures and Measure Groups L04-1

Module 5 Lab: Using MDX L05-1

Module 6 Lab: Customizing a Cube L06-1

Module 7 Lab: Implementing a Tabular Data Model in SQL Server Analysis

Services L07-1

Module 8 Lab: Using DAX to Enhance a Tabular Data Model L08-1

Module 9 Lab: Using Data Mining to Support a Marketing Campaign L09-1

 MCT USE ONLY. STUDENT USE PROHIBITED
About This Course i

About This Course

This section provides a brief description of the course, audience, suggested prerequisites, and course
objectives.

Course Description
The focus of this 3-day instructor-led course is on creating managed enterprise BI solutions. It describes
how to implement both multidimensional and tabular data models and how to create cubes, dimensions,
measures, and measure groups.

Audience
The primary audience for this course are database professionals who need to fulfil BI Developer role to
create enterprise BI solutions. Primary responsibilities will include:

 Implementing multidimensional databases by using SQL Server Analysis Services

 Creating tabular semantic data models for analysis by using SQL Server Analysis Services

The secondary audiences for this course are ‘power’ information workers.

Student Prerequisites
In addition to their professional experience, students who attend this training should already have the
following technical knowledge:

 At least 2 years’ experience of working with relational databases, including:

o Designing a normalized database

o Creating tables and relationships

o Querying with Transact-SQL

o Some basic knowledge of data warehouse schema topology (including star and snowflake
schemas)
o Some exposure to basic programming constructs (such as looping and branching)

o An awareness of key business priorities such as revenue, profitability, and financial accounting is
desirable

Course Objectives
After completing this course, students will be able to:
 Describe the components, architecture, and nature of a BI solution.

 Create a multidimensional database with analysis services.

 Implement dimensions in a cube.

 Implement measures and measure groups in a cube.

 Use MDX syntax.

 Customize a cube.

 Implement a tabular database.

 Use DAX to query a tabular model.

 Use data mining for predictive analysis.

 MCT USE ONLY. STUDENT USE PROHIBITED
ii About This Course

Course Outline
The course outline is as follows:

 Module 1: ‘Introduction to Business Intelligence and data modelling’ introduces key BI concepts and
the Microsoft BI product suite

 Module 2: ‘Creating multidimensional databases’ describes the steps required to create a

multidimensional database with analysis services.

 Module 3: ‘Working with cubes and dimensions’ describes how to implement dimensions in a cube.
 Module 4: ‘Working with measures and measure groups’ describes how to implement measures and
measure groups in a cube.

 Module 5: ‘Introduction to MDX’ describes the MDX syntax and how to use MDX.

 Module 6: ‘Customizing cube functionality’ describes how to customize a cube.

 Module 7: ‘Implementing a tabular data model by using analysis services’ describes how to
implement a tabular data model in PowerPivot.
 Module 8: ‘Introduction to Data Analysis Expression (DAX)’ describes how to use DAX to create
measures and calculated columns in a tabular data model.
 Module 9: ‘Performing predictive analysis with data mining’ describes how to use data mining for
predictive analysis.

Course Materials
The following materials are included with your kit:
 Course Handbook: a succinct classroom learning guide that provides the critical technical
information in a crisp, tightly-focused format, which is essential for an effective in-class learning
experience.

o Lessons: guide you through the learning objectives and provide the key points that are critical to
the success of the in-class learning experience.

o Labs: provide a real-world, hands-on platform for you to apply the knowledge and skills learned
in the module.

o Module Reviews and Takeaways: provide on-the-job reference material to boost knowledge
and skills retention.

o Lab Answer Keys: provide step-by-step lab solution guidance.

Additional Reading: Course Companion Content on the

https://www.microsoft.com/en-us/learning/companion-moc.aspx Site: searchable, easy-to-
browse digital content with integrated premium online resources that supplement the Course
Handbook.

 Modules: include companion content, such as questions and answers, detailed demo steps and
additional reading links, for each lesson. Additionally, they include Lab Review questions and answers
and Module Reviews and Takeaways sections, which contain the review questions and answers, best
practices, common issues and troubleshooting tips with answers, and real-world issues and scenarios
with answers.
 MCT USE ONLY. STUDENT USE PROHIBITED
About This Course iii

 Resources: include well-categorized additional resources that give you immediate access to the most
current premium content on TechNet, MSDN®, or Microsoft® Press®.

Additional Reading: Student Course files on the https://www.microsoft.com/en-

us/learning/companion-moc.aspx Site: includes the Allfiles.exe, a self-extracting executable
file that contains all required files for the labs and demonstrations.

 Course evaluation: at the end of the course, you will have the opportunity to complete an online
evaluation to provide feedback on the course, training facility, and instructor.

 To provide additional comments or feedback on the course, send email to [email protected].

To inquire about the Microsoft Certification Program, send an email to [email protected].

Virtual Machine Environment

This section provides the information for setting up the classroom environment to support the business
scenario of the course.

Virtual Machine Configuration

In this course, you will use Microsoft® Hyper-V® to perform the labs.

Note: At the end of each lab, you must revert the virtual machines to a snapshot. You can
find the instructions for this procedure at the end of each lab

The following table shows the role of each virtual machine that is used in this course:

Virtual machine Role

20768C-MIA-DC MIA-DC1 is a domain controller.

20768C-MIA-SQL MIA-SQL has SQL Server 2017 installed

MSL-TMG1 TMG1 is used to access the internet

Software Configuration
The following software is installed on the virtual machines:

 Windows Server 2016

 SQL2017

 Microsoft Office 2016

 SharePoint 2016

Course Files
The files associated with the labs in this course are located in the D:\Labfiles folder on the 20768B-MIA-
SQL virtual machine.

Classroom Setup
Each classroom computer will have the same virtual machine configured in the same way.
 MCT USE ONLY. STUDENT USE PROHIBITED
iv About This Course

Course Hardware Level

To ensure a satisfactory student experience, Microsoft Learning requires a minimum equipment
configuration for trainer and student computers in all Microsoft Learning Partner classrooms in which
Official Microsoft Learning Product courseware is taught.

 Intel Virtualization Technology (Intel VT) or AMD Virtualization (AMD-V) processor

 Dual 120-gigabyte (GB) hard disks 7200 RM Serial ATA (SATA) or better

 16 GB of random access memory (RAM)

 DVD drive

 Network adapter

 Super VGA (SVGA) 17-inch monitor

 Microsoft mouse or compatible pointing device

 Sound card with amplified speakers

Additionally, the instructor’s computer must be connected to a projection display device that supports
SVGA 1024×768 pixels, 16-bit colors.
 MCT USE ONLY. STUDENT USE PROHIBITED
1-1

Module 1
Introduction to Business Intelligence and Data Modeling
Contents:
Module Overview 1-1 
Lesson 1: Overview of Business Intelligence and Data Models 1-2 

Lesson 2: Microsoft Business Intelligence Platform 1-7 

Lab: Exploring a Business Intelligence Solution 1-14 

Module Review and Takeaways 1-18 

Module Overview
Business intelligence (BI) is a fundamental part of everyday business activity for modern organizations,
enabling them to understand trends and plan future activities more effectively. A great deal of BI depends
on data models, which expose data in a meaningful way, and enable much faster analysis of complex data.

Objectives
After completing this module, you will be able to:

 Describe business intelligence and data modeling.

 Describe the technologies that make up the Microsoft business intelligence platform.
 MCT USE ONLY. STUDENT USE PROHIBITED
1-2 Introduction to Business Intelligence and Data Modeling

Lesson 1
Overview of Business Intelligence and Data Models
The term “business intelligence” is commonly used to refer to a broad range of activities and technologies
that enable organizations to better understand their data, and to use it to gain insights that can benefit
the business. Many organizations use data models as the basis of their BI activities because they can
improve usability and offer other benefits such as improved performance. Any BI project is a substantial
undertaking and requires careful planning. To ensure that projects are successful and achieve the required
return on investment (ROI), it is important to have an understanding of the key project roles in this type of
project.

Lesson Objectives
After completing this lesson, you will be able to:

 Describe key business intelligence scenarios.

 Describe current trends in business intelligence.

 Describe the purpose of data models, and their benefits.

 Explain the components of business intelligence projects and the key project roles.

Business Intelligence Scenarios

Businesses today have access to rapidly increasing
volumes of data, which business users must use to
make ever-faster decisions in day-to-day business
operations. A typical BI solution enables end users,
such as data analysts and other information
workers, to produce reports and associated
documentation, to view this kind of information,
and to perform sophisticated data analysis. Decision
makers can use the information in reports, and the
insights that are obtained from data analysis, to
better understand past performance and to guide
them in their ongoing planning.

Reporting
Reporting is the communication of information that is gained from BI. Most organizations rely on reports
to summarize business performance and activities. Consequently, most BI solutions include an element
that generates reports. Typical examples are financial and management reports that include cash flow,
profit and loss, a balance sheet and open orders. A retail business might require stock inventory reports,
whereas a technical support call center might need a report that shows call log data. In some scenarios,
users might need to view reports interactively in a web browser or custom application. In other instances,
the requirement might be to send reports as email attachments in specific formats, such as Microsoft®
Excel® workbooks or Microsoft Word documents. In many cases, the reports might need to be printed,
for example, to send a physical report to customers or shareholders. When you are planning a reporting
solution, you must consider the reports that are required, the audiences for those reports, and how they
will be delivered. Microsoft SQL Server® includes Reporting Services, a powerful reporting technology
that supports a wide range of reporting scenarios.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 1-3

Note: This course does not describe SQL Server Reporting Services in detail. For more
information about using Reporting Services, you can attend course number 10990C: Analyzing
Data with SQL Server Reporting Services.

Analysis
Analysis is the interpretation of business data that the BI solution delivers. For business analysts in
particular, performing analysis is a discrete activity that involves using specialist analytical tools to examine
data in analytical models. For others, analysis is simply a part of everyday work and involves using reports
or dashboards as a basis for business decision making. In general, when you are planning a BI solution,
you should consider the following kinds of analytical requirements:
 Interactive analysis. Some BI solutions must support interactive “slice and dice” analysis in business
tools such as Excel or specialist data analysis tools. The resulting information can then be published as
a report.
 Dashboards and scorecards. Commonly, analytical data can be summarized in a dashboard or
scorecard and embedded into business applications or portals, such as Microsoft SharePoint® Server
sites. These solutions might provide some interactivity to enable users to “drill down” into specific
details, or they may simply show important key performance indicators (KPIs).

 Data mining. Most analysis and reporting concerns historical data, but a BI solution can also support
predictive analysis by using historical data to determine trends and patterns.

Trends in Business Intelligence

There have been many changes in the area of BI in
recent years, and the traditional model of a
complex, on-site, IT-managed BI solution is no
longer the only option for organizations.

Broader Adoption of Business

Intelligence
BI was once the preserve of large companies that
had the financial resources to fund complex
projects, and the in-house knowledge to plan,
implement, and maintain the required technologies.

As BI technologies have become much less

expensive and more straightforward to implement, more and more organizations have started to use
them. Furthermore, the availability of BI solutions in cloud services such as Microsoft Azure® has lowered
the barrier to entry for smaller organizations by removing the need to purchase and maintain expensive
hardware.

Self-Service Business Intelligence

Traditionally, information workers have had to rely on IT departments and technical users not only to
manage the infrastructure, but also to oversee the creation of reports. However, in recent years, the trend
has been to move away from this “IT-managed model” toward a “self-service BI model,” which enables
information workers to directly create and manage the reports that they need. Business users are
increasingly familiar with technology and adept at using it to perform their jobs. Although complex report
authoring and delivery remains a task that is best suited to IT and BI professionals, many organizations are
adopting self-service reporting solutions that take advantage of the available business data and IT
 MCT USE ONLY. STUDENT USE PROHIBITED
1-4 Introduction to Business Intelligence and Data Modeling

sophistication of users. This self-service reporting approach is of significant benefit in scenarios where
businesses need to:

 Empower information workers to get the information that they need to make informed decisions
without waiting for an IT professional to create a report.

 Supplement standard reports that the IT department create with custom reports for specific job roles
or scenarios.

 Reduce IT workload by minimizing requests for custom reports, enabling IT professionals to focus on
more strategic and operational tasks.

 Promote collaboration—when information workers can easily share the reports and other documents
that they create, collaboration becomes much easier.
For self-service reporting to be effective, some initial work must be done to design and implement a
suitable reporting infrastructure. After that is in place, users can benefit from the ability to customize the
reports that they use without placing an additional burden on the IT department.

Out-of-the-Box Solutions
Another significant trend in BI is the availability of out-of-the-box solutions, which are all-in-one offerings
that can accelerate the time that it takes to plan and implement a BI solution. Reducing planning and
implementation times makes it possible to achieve a much faster ROI, but the upfront cost of these
systems can be a barrier for some organizations.

Introduction to Data Models

Most corporate BI solutions include analytical data
models that provide information workers with a
way to “slice and dice” data warehouse measures by
aggregating them across the dimensions. Often,
these analytical models are referred to as “cubes.”
Technically, a cube is a specific organization of
measures and dimensions in a multidimensional
data model, but the word “cube” is commonly used
as a generic term referring to any data model that
enables users to aggregate measures by business
entities.

Benefits of Analytical Data Models

It is possible to create BI solutions that support reporting and analysis directly from tables or views in a
database or data warehouse. However, in most scenarios, creating a separate analytical data model layer
results in the following benefits:

 The data model abstracts the underlying data warehouse tables, which enables you to create models
that reflect how business users perceive the business entities and measures, regardless of the data
warehouse table schema. If necessary, you can modify or expand the underlying data warehouse
without affecting the data model that is used by business users for analysis.

 The data model reflects the users’ view of the business, so data analysis is easier for information
workers who have little or no understanding of database schema design. You can use meaningful
names for tables and fields, and define hierarchies based on attributes in dimension tables that make
the data more intuitive for business users.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 1-5

 You can add custom logic to a data model that increases business value when analyzing the data. For
example, you can define KPIs that make it easier to compare actual business measures with targets.

 Although the SQL Server database engine can provide extremely fast query performance, a data
warehouse typically contains a massive volume of data. Most analyses involve aggregating measures
across multiple dimensions, so the processing overhead for complex queries can result in
unacceptably slow response times, especially when many users access the data concurrently. A data
model typically preaggregates the data, which provides vastly superior performance for analytical
queries.

 Data models are a common feature in BI solutions and there are well-established standards. By
creating a data model, you can expose analytical data through a standard interface to be consumed
by client applications such as Excel or third-party analytical tools.

Note: In addition to using data models as a source for reporting and analysis, you can also
create analytical reports directly from the data warehouse or a departmental data mart. This
enables you to express queries by using Transact-SQL, which may be more familiar to a report
developer than a data modeling query language such as Multidimensional Expressions (MDX) or
Data Analysis Expressions (DAX).

Business Intelligence Projects

A data model is only part of a BI project. Other key
components of a BI project might include:

 A data warehouse. A data warehouse is a

database that acts as the central repository for
the data that you use in analysis and reporting.
Typically, a data warehouse has a specialized
schema, such as a star or snowflake schema,
and contains two types of data. There is “fact”
data, which is typically the numerical data—
such as sales data—that you want to analyze,
and “dimension” data—such as date or
customer data—which you use to filter the fact
data.

 Extract, Transform, and Load (ETL) processes. ETL is the process of extracting data from data
sources, applying transformations to convert the data into the required format, and then loading it
into a data store such as a data warehouse. Source data can come from many sources, and might be
in various formats, and it is usually necessary to perform regular data loads to keep a data warehouse
up to date.

 Reporting and analysis tools. There are many different tools that you can use to create reports and
analyze data, including Excel, Reporting Services, and Report Builder. The tools that end users will
require can have a big influence on the design of the overall solution.
 MCT USE ONLY. STUDENT USE PROHIBITED
1-6 Introduction to Business Intelligence and Data Modeling

Project Roles
Implementing a BI project can be a complex process, and typically requires the involvement of people
from across the business. Most BI projects include some or all of the roles that are described below (note
that the project role names might differ; it is the function of the role that is important):

 A project manager. Coordinates project tasks and schedules, ensuring that the project is completed
on time and within budget.

 A BI solution architect. Has overall responsibility for the technical design of the data warehousing
solution.

 A data modeler. Designs the data warehouse schema and analytical data models.

 A database administrator. Designs the physical architecture and configuration of the data
warehouse database. Database administrators who have responsibility for data sources that are used
in the data warehousing solution must also be involved in the project to provide access to the data
sources that the ETL process uses.

 An infrastructure specialist. Implements the server and network infrastructure for the data
warehousing solution.

 An ETL developer. Builds the ETL workflow for the data warehousing solution.

 A report developer. Creates the reporting elements of the BI solution.

 Business users. Provide requirements and help to prioritize the business questions that the data
warehousing solution will answer. Often, the team includes a business analyst as a full-time member
to help interpret the business questions and ensure that the solution design meets users’ needs.

 Testers. Verify the business and operational functionality of the solution as it is developed.

Note: The list in this topic is not exhaustive, but represents roles that must be performed. In
some cases, multiple roles may be performed by one person, although, in general, you should
avoid having testers validate their own development work.

Verify the correctness of the statement by placing a mark in the column to the right.

Statement Answer

True or false? A data model, which

is created by using a technology
such as SQL Server Analysis
Services, is a required component
of all BI projects.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 1-7

Lesson 2
Microsoft Business Intelligence Platform
Microsoft products are used to provide IT infrastructure for organizations all over the world. Therefore, it
makes sense for many of these organizations to consider using the Microsoft BI platform to benefit from
the close integration and common infrastructure capabilities of the various products that can help to
deliver an enterprise BI solution. As a Microsoft BI professional, you need to know which products can be
used to implement the various elements of a BI solution, and how those products integrate with each
other.

Lesson Objectives
After completing this lesson, you will be able to describe:

 The role of SQL Server in a BI solution.

 The role of SharePoint Server in a BI solution.

 The role of Microsoft Office applications in a BI solution.

 The role of Power BI Desktop in a BI solution.

 The types of analytical data models that SQL Server Analysis Services supports.

 The installation options for SQL Server Analysis Services.

 Some considerations for SharePoint integration.

SQL Server
SQL Server provides the core data services for a BI
solution. These services include:

 The SQL Server database engine, which is used

for application databases, operations
databases, and the data warehouse throughout
the BI solution.

 SQL Server Integration Services, which is used

as the primary platform for ETL processes.

 Data Quality Services, which provides data

cleansing and matching capabilities.

 Master Data Services, which provides master

data management capabilities.

 SQL Server Analysis Services, which provides a storage and query processing engine for
multidimensional and tabular data models.

 SQL Server Reporting Services, which provides a platform for publishing and delivering reports that
users can consume through a native web-based interface or have delivered by subscriptions.
 MCT USE ONLY. STUDENT USE PROHIBITED
1-8 Introduction to Business Intelligence and Data Modeling

SQL Server Editions

SQL Server is available in the following core editions:

 SQL Server Enterprise. You should use this edition for data warehouses and BI solutions that require
advanced SQL Server Integration Services features, such as fuzzy logic and change data capture (CDC)
components.

 SQL Server Standard. You can use this edition for solutions that require basic reporting in SQL Server
Reporting Services, multidimensional models in SQL Server Analysis Services, and basic data mining.

Note: SQL Server is also available in Web and Express editions, but these are generally not
appropriate for BI solutions. SQL Server Developer edition includes all the functionality of
Enterprise edition, but is only licensed for development and test systems.

SharePoint Server
SharePoint Server provides enterprise information-
sharing services through collaborative websites.
SharePoint Server provides the following BI
capabilities:

 Excel Services. Users can view and interact

with Excel workbooks that are shared in a
SharePoint document library through a web
browser. This includes workbooks that use data
connections to query data in a data warehouse
or SQL Server Analysis Services data model.
 PowerPivot Services. Users can share and
interact with Excel workbooks that contain a
PowerPivot tabular data model. This enables business users to create and share their own analytical
data models.

 Integration with SQL Server Reporting Services. You can deliver and manage reports and data
alerts through SharePoint document libraries instead of the native Report Manager interface that SQL
Server Reporting Services provides.

 Power View. Power View is an interactive data visualization technology through which users can
graphically explore a data model in a web browser.
 PerformancePoint Services. PerformancePoint Services enables BI developers to create dashboards
and scorecards that deliver KPIs and reports through a SharePoint site.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 1-9

Office Applications
Microsoft Office provides productivity applications
that business users can use to consume and interact
with BI data. These applications include:

 Microsoft Excel. Excel is the most commonly

used data analysis tool in the world, and can be
used to:
o Import data from a data warehouse and
use it to create charts and reports.
o Create interactive PivotTables and
PivotCharts from analytical data models in
SQL Server Analysis Services or PowerPivot.

o Create PowerPivot workbooks that contain tabular data models.

o Create Power View visualizations from tabular models.

 Microsoft Word. Word is a document-authoring tool. In a BI scenario, users can export SQL Server
Reporting Services reports in Word format and use the editing and reviewing tools in Word to
enhance them.

 Microsoft PowerPoint®. PowerPoint is a widely used presentation tool. Users can save Power View
visualizations as PowerPoint presentations, and present business data in a dynamic, interactive format.
 Microsoft Visio®. Visio is a diagramming tool that can be used to document database schemas and
to visualize data mining analyses.

Power BI Desktop
Business users can also use Power BI Desktop to
consume and interact with BI data. It shares many
of the features offered by the Excel power tools, so
business users will find transitioning between the
two tools to be a straightforward process. Power BI
Desktop is a stand-alone tool that enables you to:

 Import data. You can connect to a wide range

of data sources including files, databases,
Azure, and online services.

 Transform data. The Query Editor includes the

same functionality as Get & Transform in Excel.
You can also use DAX to shape your data and
create calculated tables and columns.
 Create reports. After importing, and transforming your data, you can drag visuals or fields onto the
report designer, to begin building reports. The visuals can be customized with colors, titles and text,
and other settings applicable to each type of chart. You can also create or download custom visuals
for your reports.
 MCT USE ONLY. STUDENT USE PROHIBITED
1-10 Introduction to Business Intelligence and Data Modeling

 Publish reports. Using the Power BI Desktop application, you can publish reports to the online
Power BI service. If you need to keep your data inside your organization, you can use Power BI
Desktop (Optimized for Power BI Report Server) to publish reports to an on-premise Power BI Report
Server.
 Create dashboards. After publishing a report, you can use the report items, known as tiles, to create
a new dashboard, even combining charts, maps, and KPIs from different reports.

Power BI Mobile
Power BI offers a mobile app for iOS, Android, and Windows devices. Reports and dashboards
automatically adjust their size to fit the screen of the device, so you need not worry about creating mobile
versions of your work. The apps are free to download, and reports and dashboards are fully interactive.

Q&A

One very useful feature of Power BI is Q&A. This means you can ask questions of your data using the
Natural Query Language. You can type in a topic, such as Total sales last year in Canada. You can also
specify which chart visual the data should be presented in. When Power BI returns the result, you can pin
the visual to a new or existing dashboard.

Analytical Data Models

SQL Server includes two types of data model,
multidimensional data models and tabular data
models:
 Multidimensional data models. You can use a
multidimensional data model to create a SQL
Server Analysis Services database that contains
one or more cubes, each providing
aggregations of measures in measure groups
across multiple dimensions.
 Tabular data models. For a user who is
performing analysis, tabular models provide
similar functionality to a multidimensional
model. In many cases, the two models are indistinguishable from one another. However, for BI
developers, tabular models are often faster and easier to create because they do not require as much
online analytical processing (OLAP) modeling knowledge as multidimensional models. Tabular data
models are based on relationships between multiple tables of data, so they are much easier to
understand for professionals who are used to working with relational databases.

Note: In SQL Server, analytical data models are sometimes also called BI Semantic Models.

Selecting a Data Model

Many factors are involved in selecting a data model for a specific BI scenario. As you learn about the data
models in this course, you will understand more about the differences between the two types of model,
and when you might choose to use them. In general terms, the following factors will influence the choice
of data model:

 Hardware. Tabular data models use in-memory storage, so they require relatively large amounts of
memory.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 1-11

 Calculations. Data models usually include calculations—you can implement these in tabular data
models by using the DAX language and in multidimensional models by using the MDX language.
Some calculations are easier to implement in DAX, others in MDX.

 Many-to-many relationships. Tabular data models do not support the creation of many-to-many
relationships, although you can use DAX to model this type of relationship.

 Data sources. Tabular data models support a wider range of data sources than multidimensional
data models, including text files and Excel.

 Available skills. As with many technical decisions, a frequent deciding factor is the knowledge and
skills of the developers who will implement the data model.

 Client tools. Most Microsoft tools can consume data from either type of data model.

Installation Options for SQL Server Analysis Services

SQL Server Analysis Services is a feature of SQL
Server that you install separately from the SQL
Server database engine. When you install SQL
Server Analysis Services, you must select the mode
in which the SQL Server Analysis Services instance
will run. There are three installation modes:

 Multidimensional and data mining mode.

This mode enables the creation of
multidimensional data models, and additionally
includes data mining functionality. Data mining
is the process of analyzing data to identify
trends, relationships, and patterns, and using
this to predict future behavior.

 Tabular mode. Tabular mode enables the creation of tabular data models.
 PowerPivot mode. PowerPivot mode, which is also known as SharePoint integrated mode, integrates
SQL Server Analysis Services with a SharePoint server instance to make features of SQL Server Analysis
Services available through a SharePoint portal.

You cannot install two modes of SQL Server Analysis Services as part of the same instance, and you cannot
switch modes after installation. For example, if you need to create both tabular data models and
multidimensional data models, you should install two separate instances, each of which uses the
appropriate mode. As with instances of the database engine, it is possible to install multiple instances of
SQL Server Analysis Services on the same server running Windows Server®.
 MCT USE ONLY. STUDENT USE PROHIBITED
1-12 Introduction to Business Intelligence and Data Modeling

Considerations for SharePoint Integration

SharePoint Server enables the sharing and
collaboration of reports, and is an important
component of many BI solutions. By installing SQL
Server Analysis Services in PowerPivot mode, you
can take advantage of SharePoint to provide
multiple benefits, including:

 A single location for PowerPivot reports, which

users access through the easy-to-use
PowerPivot Gallery.

 Reports remain within the scope of

management—for example, by being included
in backup runs. The PowerPivot Management
Dashboard provides a single interface that administrators can use to manage the PowerPivot mode
deployment—for example, by monitoring resource usage, how often reports are used, and data
refresh successes and failures.

SharePoint Architecture
When planning SharePoint integration, there are three tiers of the SharePoint architecture to consider:
 Web front-end tier. One or more servers that accept requests for a SharePoint application or service,
and direct the request to the appropriate application server.

 Application tier. One or more servers that host the service applications in the SharePoint Server
infrastructure.
 Data tier. A SQL Server instance that hosts SharePoint databases. It is common to use clustering to
ensure high availability of the SharePoint databases.

SharePoint Business Intelligence Topologies

By changing where you host each tier, you can deploy SharePoint in one of three topologies:

 Single server. In this topology, you host all of the SharePoint Server architecture layers on a single
server running Windows Server. The main benefit of this model is that it minimizes the licensing cost
for the solution. Typically, you use this type of configuration in development environments or training
environments because it is the easiest way to set up a SharePoint farm. However, because all three
layers run on the same server running Windows Server and share the same hardware, there can be
increased contention for resources. If this affects the performance of business reports, consider
implementing a scale-out solution.

 Scale-out. In a scale-out topology, you separate each of the SharePoint Server architecture layers
onto different servers running Windows Server. Scaling out a SharePoint farm distributes the
workload across multiple servers, which reduces resource contention and improves throughput.
However, this topology comes with an additional licensing cost because it uses multiple servers.
Furthermore, it requires greater preparation to ensure that security works seamlessly across the
SharePoint farm. This topology offers no resilience if one of the servers shuts down. You should
consider a scale-out topology if performance is important, but there is no requirement for resilience.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 1-13

 High availability. The high-availability topology involves separating the SharePoint architecture
layers across servers running Windows Server, and then duplicating each layer on another server. This
is the most expensive topology to implement, but it provides load balancing and high availability
across the entire SharePoint farm. The infrastructure preparation is similar to that of creating a scale-
out architecture, and also requires a network load balancer to distribute incoming requests to the first
available web front-end server.

Installing SQL Server Analysis Services in PowerPivot Mode

After setting up the SharePoint farm, you can use the SQL Server installation media to install PowerPivot
for SharePoint on the application servers in the farm. This creates a PowerPivot mode instance of SQL
Server Analysis Services on the application server, which provides a workspace for working with
PowerPivot files. After the installation is complete, you can download the PowerPivot for SharePoint add-
in file spPowerPivot.msi, and install it on each server in the SharePoint Server farm. The PowerPivot for
SharePoint add-in installs data providers, the PowerPivot configuration tool, the PowerPivot Gallery, and
components that are required for data refresh.

Check Your Knowledge

Question

You are part of a team that is in the early

stages of planning a BI project. You have
identified that a key requirement for the
project will be the ability to use historical
data to make predictions about future
scenarios, which the marketing
department will use to help to create new
campaigns. Which type of SQL Server
Analysis Services instance will you need to
include so that you can meet this
requirement?

Select the correct answer.

An instance of SQL Server Analysis

Services in tabular mode.

An instance of SQL Server Analysis

Services in multidimensional and data
mining mode.

An instance of SQL Server Analysis

Services in PowerPivot mode.
 MCT USE ONLY. STUDENT USE PROHIBITED
1-14 Introduction to Business Intelligence and Data Modeling

Lab: Exploring a Business Intelligence Solution

Scenario
Adventure Works Cycles is a multinational corporation that manufactures and sells bicycles and cycling
accessories. The company sells its products through an international network of resellers and, in recent
years, has developed a direct sales channel through an e-commerce website.

The company is financially sound and has a strong order book, but sales volumes have remained relatively
static for the past few years. Senior management is under pressure from shareholders to develop a growth
strategy that will drive increased revenue and profit. Management believes that a key factor in its growth
strategy is investment in technology that improves collaboration between the various divisions of the
company, and enables them to track and share key business performance metrics.

Objectives
After completing this lab, you will have:

 Explored a data warehouse by using a database diagram and by issuing Transact-SQL queries.
 Performed basic data analysis by using Excel with a data model as a data source.

Estimated Time: 45 minutes

Virtual machine: 20768C-MIA-SQL

User name: ADVENTUREWORKS\Student

Password: Pa55w.rd

Exercise 1: Exploring a Data Warehouse

Scenario
Adventure Works Cycles has a data warehouse containing data that relates to Internet and reseller sales.
This data warehouse will form the foundation for an enterprise BI solution to include SQL Server Analysis
Services data models and a SQL Server Reporting Services solution for reports.

The main tasks for this exercise are as follows:

1. Prepare the Lab Environment

2. Explore a Data Warehouse Schema

3. Query a Data Warehouse

 Task 1: Prepare the Lab Environment

1. Ensure that the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and then
log on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd.

2. Run Setup.cmd in the D:\Labfiles\Lab01\Starter folder as Administrator.

 Task 2: Explore a Data Warehouse Schema

1. Use SQL Server Management Studio to connect to the MIA-SQL SQL Server instance.

2. Create a database diagram for the AdventureWorksDW database, and note the following details
about its schema:

o The FactInternetSales table stores details of sales orders that are made through the Adventure
Works website. The table is related to the DimCustomer table, which contains details of the
customers who have placed orders.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 1-15

o The FactResellerSales table stores details of sales orders that are made to resellers. The table is
related to the DimReseller table, which contains details of the resellers who have placed orders.
o The DimDate table stores values for dates, including details of the calendar and fiscal periods in
which individual dates occur.

o The FactInternetSales and FactResellerSales tables are related to the DimDate table by
multiple fields. These fields represent the order date (the date when the order was placed), the
due date (the date when the order was expected to be in stock), and the ship date (the date
when the order was shipped to the customer).
o Both the DimCustomer and DimReseller tables are related to the DimGeography table, which
stores details of geographical locations. The DimSalesTerritory table is also related to
DimGeography.

o The DimProduct table contains details of products, and is related to the

DimProductSubcategory table, which is in turn related to the DimProductCategory table.

o Many tables include multiple language values for the same data value, for example, the DimDate
table stores the English, French, and Spanish words for each month name.

o The DimEmployee table is related to itself to represent the fact that each employee has a
manager, who is also an employee.
3. Save the diagram as AdventureWorksDW Schema.

 Task 3: Query a Data Warehouse

1. Open the Query DW.sql script file in the D:\Labfiles\Lab01\Starter folder.
2. Execute the query under the comment Internet sales by year and month, and note the following
details:

o The data warehouse contains historical Internet sales orders from July 2005 to July 2008. Reseller
sales in the data warehouse are also recorded for this time period.
o You can use the MonthNumberOfYear column in the DimDate table to sort month names into
chronological order—without this field, it would be difficult (though not impossible) for reporting
clients to sort months other than alphabetically. A similar field named DayNumberOfWeek can
be used to sort weekday names into chronological order.

3. Execute the query under the comment Geographical reseller sales, and note the following details:

o In 2005, Adventure Works only sold to resellers in the United States and Canada.

o In 2006, this was expanded to include France and the United Kingdom.

o In 2007, resellers in Australia and Germany were added.

o By contrast, Adventure Works has sold directly to Internet customers in all of these regions since
2005.

4. Execute the query under the comment Sales by product category, and note the following details:

o Adventure Works sells four categories of product: Accessories, Bikes, Clothing, and Components.
o Components are only sold to resellers, not to Internet customers.

o Accessories were not sold to Internet customers until 2007.

 MCT USE ONLY. STUDENT USE PROHIBITED
1-16 Introduction to Business Intelligence and Data Modeling

Results: After completing this exercise, you will have:

Used a database diagram to explore a data warehouse schema.

Used Transact-SQL queries to explore the data in a data warehouse.

Exercise 2: Exploring a Data Model

Scenario
Having examined the data warehouse, you will now explore a SQL Server Analysis Services data model
that is based on the data warehouse tables.

The main tasks for this exercise are as follows:

1. View a SQL Server Analysis Services Database

2. Create a PivotTable in Excel

3. Filter the PivotTable

 Task 1: View a SQL Server Analysis Services Database

1. In SQL Server Management Studio, connect to the MIA-SQL instance of SQL Server Analysis Services.

2. Explore the Adventure Works OLAP database in the MIA-SQL SQL Server Analysis Services server.
Note that it contains a data source named Adventure Works Data Warehouse, which connects to
the AdventureWorksDW database that you explored in the previous exercise.

3. Browse the cube named Internet Sales, to view the Sales Amount measure by Order Date – Fiscal
Year. Note that in Adventure Works, fiscal years run from July to June, so fiscal year 2006 represents
the period from July 2005 to June 2006.

4. Close SQL Server Management Studio when you are finished.

 Task 2: Create a PivotTable in Excel

1. Create a new blank Excel workbook.
2. Connect to the MIA-SQL server by using its fully qualified domain name, MIA-
SQL.adventureworks.msft. With a user name of ADVENTUREWORKS\Student.

3. Import data from the MIA-SQL SQL Server Analysis Services server:

o Select the Internet Sales cube in the Adventure Works OLAP database.

o Import the data into a PivotTable report.

4. In the PivotTable, display the Sales Amount measure. Change the settings of this value field to
format it by using the Accounting format.

5. Add the Order Date.Calendar Date hierarchy to the rows of the PivotTable, and explore the data by
drilling down from years to semesters, semesters to quarters, quarters to months, and months to days.

6. Save the workbook as Sales.xlsx in the D:\Labfiles\Lab01\Starter folder.

 Task 3: Filter the PivotTable

1. Add a slicer based on the English Product Category Name field in the Product Category hierarchy
to the PivotTable that you created in the previous task:
o Note that the Components category is disabled because there are no sales of components to
Internet customers.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 1-17

2. Use the filter to show only sales for the Accessories category:

o Note that no accessories were sold in 2005 or 2006.

3. Clear the filter, and then save and close the workbook.

Results: After completing this exercise, you will have used Excel to explore an analytical data model built
on the data warehouse.

Question: Which tools have you used to view BI data? How do these tools differ and what
are the advantages and disadvantages of these tools?
 MCT USE ONLY. STUDENT USE PROHIBITED
1-18 Introduction to Business Intelligence and Data Modeling

Module Review and Takeaways

In this module, you have learned about the common elements of a BI solution and how Microsoft
software products can help you to implement them. You have also learned about the key roles and
scoping tasks in a BI project.

Review Question(s)
Question: In your experience, what are the factors that affect the success or failure of a BI
project?
 MCT USE ONLY. STUDENT USE PROHIBITED
2-1

Module 2
Creating Multidimensional Databases
Contents:
Module Overview 2-1 
Lesson 1: Introduction to Multidimensional Analysis 2-2 

Lesson 2: Data Sources and Data Source Views 2-9 

Lesson 3: Cubes 2-13 

Lesson 4: Overview of Cube Security 2-19 

Lesson 5: Configuring SSAS 2-25 

Lesson 6: Monitoring SSAS 2-30 

Lab: Creating a Multidimensional Database 2-35 

Module Review and Takeaways 2-41 

Module Overview
You can use Microsoft® SQL Server® Analysis Services (SSAS) to build on Online Analytical Processing
(OLAP) solutions that help you to quickly and efficiently obtain accurate business intelligence (BI) that
improves business decision making—and helps to drive growth. SSAS multidimensional databases, which
are also known as OLAP cubes, have several benefits including:
 Fast query response times because of a preaggregation of data and calculations. Queries against
cubes are generally much quicker than the equivalent queries against relational databases.

 Multidimensional structures are logical and straightforward, and also provide an intuitive navigation
path for end users that is easy to understand.

 Accessibility from a wide range of tools, such as Microsoft Excel®, Power View, and third-party tools.

This module introduces multidimensional databases and the core components of an OLAP cube.

Objectives
After completing this module, you will be able to:

 Describe the core components of a multidimensional database solution.

 Create data sources and data source views.

 Create and browse a multidimensional cube.

 Explain how to implement security in a multidimensional database.

 Configure memory and storage to optimize Analysis Services instances.

 Monitor Analysis Services for performance issues.

 MCT USE ONLY. STUDENT USE PROHIBITED
2-2 Creating Multidimensional Databases

Lesson 1
Introduction to Multidimensional Analysis
Multidimensional analysis using OLAP involves the creation of OLAP cubes, data sources, data source
views, dimensions, measures, and measure groups. This lesson explains the business problem that
multidimensional data models address, and describes data sources, data source views, and the core
components of a cube: dimensions, measures, and measure groups.

Lesson Objectives
After completing this lesson, you will be able to:

 Identify the business problems that multidimensional solutions can address.

 Describe the core concepts of multidimensional analysis.

 Describe the role of data sources in a multidimensional database.

 Describe the role of data source views in a multidimensional database.

 Describe the role of dimensions in a multidimensional database.

 Describe the role of measures and measure groups in a multidimensional database.

 Describe the role of cubes in a multidimensional database.

The Business Problem

Most database systems are designed for Online
Transaction Processing (OLTP). In an OLTP system,
many transactions—updates, inserts, and
deletions—can happen per second and data is
stored at the most granular level possible. This
level of detail is required and is useful for many
operations. However, in other scenarios, an OLTP
system is not always the most appropriate
solution.

For example, the fictional organization Adventure

Works sells bicycles and associated accessories.
The Adventure Works OLTP database records
every item from every sale, and these updates to the database need to happen almost instantaneously.
However, problems might emerge in situations that involve issuing complex queries, such as when the
trading manager requires a broad overview of sales for the past month by product category. The
Transact-SQL query needed to generate this data would involve filtering all sales to find only those that
occurred in the past month, grouping them by product category, and then calculating the total sales by
product group. Not only is this query potentially very slow, it will cause resource contention on the server;
it also requires the locking of tables or other objects to prevent transactions simultaneously attempting to
change the data that the query is reading. Contention and locking can negatively affect database
performance, and compromise the database’s primary role as a fast, transactional system.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 2-3

When considering using an OLTP system as a direct source of data for reporting and analysis, bear in
mind the following points:
 Aggregated spanning queries are slow to generate.

 Large queries can cause performance issues for transactional workloads.

 Normalized relational data models are not intuitive for business users.

Core Concepts for Multidimensional Data Analysis

The multidimensional data model can appear to
be quite conceptually challenging when you first
encounter it, especially if you are used to working
with relational databases. However, the basic
concepts are easy to grasp if you have a clear
understanding of the terminology.

Cubes, Measures, and Dimensions

In a multidimensional database, you create a
structure called a cube to store measures, which
are discrete, aggregated, numerical values, such as
sales revenue totals or manufacturing costs. The
cube stores measures in individual cells, and has
different dimensions by which you can analyze the data. The dimensions are analogous to the height,
width, and depth dimensions of a real, physical cube, except that they represent values such as date,
region, or product instead. Each cell is a location in the cube that represents the intersection of the
dimensions of the cube for a given set of dimension values. For example, you might select to analyze the
sales data for the product category Clothing in the South region, in the year 2016. The cell in the cube
that represents the intersection of the dimensions for the values Clothing, South, and 2016 is easily
located and contains the required value, already aggregated. The fact that data is preaggregated in a
cube means that values do not have to be calculated when a user issues a query against the cube—this is
much less resource intensive, and enables faster query response times. Cubes in multidimensional
databases nearly always have many more dimensions than the three dimensions in a real, physical cube.
This means that using the concept of a three-dimensional cube to understand multidimensional cubes is
not always intuitive!

Drilling Down, Slicing, and Dicing

You can use dimensions to drill down through different layers of data by using a hierarchy. For example,
you can initially view sales by year, then drill down to sales by month, and then by day. You can also filter
the data by using slicing and dicing dimension values:

 Slicing involves using a member of one dimension to provide a slice of the cube. For example, you
could slice a cube using a particular product, and view all sales of that product across all dates and
customers.

 Dicing involves providing values for every dimension to locate a single value for a cube. For example,
you could look for sales of a particular product, on a particular day, to a particular customer.
 MCT USE ONLY. STUDENT USE PROHIBITED
2-4 Creating Multidimensional Databases

Aggregations
One key difference between cubes and relational systems─and between cubes and tabular data models—
is the processing of the cube. The aggregations of measures are calculated when the cube is processed,
not when the query runs. Consequently, when business users run queries, response times are extremely
fast, because the complex processing is already complete.

Data Sources
A data source is an object that provides an SSAS
database with the information required to connect
to a data source for the cubes it contains. Analysis
Services can access data from one or more data
source. Typically, the source database for an OLAP
cube is a data warehouse or a departmental data
mart, in which the tables have been denormalized
to reflect a star or snowflake schema of fact and
dimension tables. However, SSAS can use virtually
any database as a data source, and abstract the
underlying data model by using a data source
view.
At a minimum, a data source includes an identifier, a name, and a connection string. The connection
string used to access the source data includes the following information:

 The type of provider, for example MSOLAP or OLE DB.

 Other properties that the provider supports or requires, such as the initial catalog (database) to
connect to, or the server name. The actual property settings for particular data source objects vary
according to the provider.
For more information about connection string properties in Analysis Services, see Connection String
Properties (Analysis Services) in Microsoft Docs:

Connection String Properties (Analysis Services)

http://aka.ms/H5sor4

Data Source Views

A data source view abstracts the underlying data
source and defines the specific tables that should
be retrieved from the data source. It can also
change the presentation of those tables and their
columns to make it more user-friendly for the end
users who will work with the data. For example,
the column names in database tables might be
meaningful to database administrators and
developers who use the database regularly, but
they might be difficult to read for other users who
are not as familiar with the database. For example,
a column name such as Orderlineitems lacks
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 2-5

space characters and capitalization, making it relatively difficult to understand. In a data source view, you
can represent columns by using names that are easier to read.
A data source view contains the following items:

 A name and a description.

 A definition of any subset of the schema retrieved from one or more data sources, which could
involve the whole schema, including the following:

o Table names

o Column names
o Data types

o Nullability

o Column lengths
o Primary keys

o Primary key/foreign key relationships

 Enhancements to the schema from the underlying data sources, including the following:
o Friendly names for tables, views, and columns.

o Named queries that return columns from one or more data sources (that show as tables in the
schema).
o Named calculations that return columns from a data source (that show as columns in tables or
views).

o Logical primary keys (needed if a primary key is not defined in the underlying table, or is not
included in the view or named query).

o Logical primary key/foreign key relationships between tables, views, and named queries.

You can also create multiple diagrams showing different subsets of the data source view. Unlike different
data source views, different diagrams reference the same schema. Therefore, any changes you make in
one diagram apply to all the others in the data source view. Diagrams are useful when you create multiple
cubes within the same solution.

Dimensions
As described earlier, dimensions provide the
context for the facts stored in the OLAP cube.
Dimensions include key values and hierarchies.

Key Values
Each dimension includes a key attribute that
matches values in the fact table. The key attribute
also enables SSAS to join records to related data in
the fact table, in the same way that tables use
columns to match values in Transact-SQL
statements. For example, a dimension called
Product Category might include a
ProductCategoryID key value, which also exists as
 MCT USE ONLY. STUDENT USE PROHIBITED
2-6 Creating Multidimensional Databases

a column in the fact table. Matching the values in the dimension and the fact table enables Analysis
Services to aggregate the measure values correctly, according to the dimension that the user specifies.

Dimension Hierarchies
Analysis Services dimensions are hierarchical, not flat. For example, a Time dimension typically has years,
months, and days; a Geography dimension typically includes country, state, and town. A dimension can
have more than one hierarchy, meaning you can drill down in different ways. For example, one hierarchy
might include year, month, and day, and another might include year, week, and day.
Dimension hierarchies can also be unbalanced and ragged. An unbalanced dimension hierarchy does not
have the same number of levels in every branch. For example, in an Employee dimension, the Sales
Director reports to the Chief Executive Officer (CEO) and has three levels of staff below her. The Executive
Assistant also reports to the CEO but has no levels below him. A ragged dimension hierarchy is a particular
type of unbalanced dimension where there are missing or duplicated levels in the hierarchy. For example,
in a Geography dimension hierarchy that contains Continent, Country or Region, State, and City, the city
state of Monaco would either have the values of Europe, Monaco, Monaco, Monaco—or the values of
Europe, Monaco, NULL, NULL.

Measures and Measure Groups

Measures are the facts aggregated at each
intersection of dimension members. They are
typically numeric and additive, and are
aggregated by using a SUM calculation, although
not all measures meet these criteria. For example,
you might use a COUNT aggregation to create a
measure that shows the number of orders placed.

Every cube includes at least one measure, but

most cubes contain many more to reflect the
different values by which an organization needs to
assess its performance. The values in measures are
frequently derived from a column in a table in a
source database, such as a fact table in a data warehouse. However, you can also create measures as
calculated members by using the Multidimensional Expressions (MDX) language. You can use this to apply
calculations to measure values or use multiple columns from a fact table in a measure expression.

Note: You will learn more about MDX in the Introduction to MDX module in this course.

Measure Groups
Measures are grouped, based on their underlying fact tables. Measure groups are used to define the
relationship between dimensions and measures. When a dimension or a measure group is added to the
cube, SSAS will examine the relationship that exists between the fact table and the dimension table in the
data source and automatically set the Measure Group information in the Dimension Usage tab of the
Cube Designer. You can then modify this to your requirements.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 2-7

Cubes
A cube is the unit that defines dimensions and
measures and is the user’s point of access to the
multidimensional data. The term “cube” is used
because it describes the multidimensional nature
of the data, as opposed to the one-dimensional or
two-dimensional types of relational tables.
However, almost all cubes have more than three
dimensions. You can use cubes to bring together
all the functionality of dimensions and measures
to display them in a user-friendly interface, such as
an Excel PivotTable table. You can use the
PivotTable interface with SSAS to display
aggregated measure data, and slice and dice it by using dimension members.

After creating a cube, you must process it. This populates the cube with data and creates the measures
and calculations specified in the cube design. Periodically, you should also process cubes to bring the data
they contain up to date, and to reflect any changes. Cubes also represent an administrative unit within
Analysis Services, and include settings that define storage, aggregation levels, and processing frequency
for the cube.
 MCT USE ONLY. STUDENT USE PROHIBITED
2-8 Creating Multidimensional Databases

Categorize Activity
Match the definition to the correct Analysis Services component.

Items

1 An object that filters a data

source and specifies a subset of
the tables to include.

2 An aggregated numerical
value, such as a sales total.

3 An object that contains data,

such as product or customer
information, that provides
context for data analysis.

Category 1 Category 2 Category 3

Data source Measure Dimension

view
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 2-9

Lesson 2
Data Sources and Data Source Views
OLAP and data mining projects in SSAS are designed based on a logical data model of related tables,
views, and queries from one or more data sources. Data sources are defined as data source objects in
Analysis Services, and logical data models are defined as data source views. You use data source views to
filter the data from a data source so that the cube includes only a subset of the data that populates a
large data warehouse.

Lesson Objectives
After completing this lesson, you will be able to:

 Create a data source.

 Create a data source view.

 Describe how to modify a data source view.

Creating a Data Source

You can use the Data Source Wizard in Visual
Studio® SQL Server Data Tools to define one or
more data sources for an Analysis Services project.
Data sources use a Connection object that defines
the connection string. In the first step of the Data
Source Wizard, you must define the Connection
object, or use an existing Connection object.

The default provider for a new connection is the

Native OLE DB\SQL Server Native Client 11.0
provider. Other supported providers include:

 .NET providers for SQLClient and OracleClient

 Native OLE DB providers for:

o Analysis Services

o Oracle

o SQL Server Native Client 11.0

If you have an existing data source defined in an SSAS database or project, and want to create a new data
source object that connects to the same underlying data source, you can copy the properties of the first
data source object into a new version.
 MCT USE ONLY. STUDENT USE PROHIBITED
2-10 Creating Multidimensional Databases

Creating a Data Source View

Data source views provide an abstraction layer
over a data source, which reduces the complexity
of the schema that you will use in Analysis
Services. You can create a data source view by
using the Data Source View Wizard in SQL Server
Data Tools. When creating a view, you must
specify the following:

 The name of the data source that the view will

use.

 The tables or views that you want to include

in the data source view.

 A name for the data source view.

SSAS design tools use data source views to maintain a cache of relational metadata, and take advantage
of some of the annotations within a data source view. Advantages of this approach include:
 A data source view makes available only tables that OLAP and data mining objects require, by
describing a subset of tables and views in a data source.

 A data source view handles the layout of tables, filters, SQL expressions, relationships, and other
complexities of the schema. Therefore, a data source view enables simple bindings by SSAS cubes,
dimensions, and mining models to the tables and columns in the data source view.

Working with Data Source Views

After you have created a data source view, you
can modify it to make the data more user-friendly.
Often, underlying data sources use naming
conventions to help developers understand the
schema. Also, the underlying data does not
typically store additional formatting information.
By using the Data Source View Designer, you can
make modifications to a data source view without
modifying the underlying source data or database
schemas. You can then present user-friendly
names and apply formatting without making any
changes to the source database systems. For
example, you can:

 Use the FriendlyName property to specify a column name from a table or view that is easier for users
to understand—or is more relevant to the subject area.
 Create a named query and named calculations that mean you can extend the relational schema of
existing tables in a data source view, without modifying the underlying data source.

 Create logical primary keys and relationships for improved performance.

 Create diagrams to reduce the visual clutter when you only want to view a subset of the tables in the
data source view.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 2-11

Demonstration: Creating a Data Source and a Data Source View

In this demonstration, you will see how to:

 Create a new Analysis Services project.

 Create a data source.

 Create a data source view.

 Modify a data source view.

Demonstration Steps
Create an Analysis Services Project

1. Ensure that the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and then
log on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd.

2. In the D:\Demofiles\Mod02 folder, run Setup.cmd as Administrator.

3. Start Visual Studio, and then click New Project.

4. In the New Project dialog box, in the Business Intelligence templates section, create a new
Analysis Services Multidimensional and Data Mining Project named Demo in the
D:\Demofiles\Mod02 folder.

Create a Data Source

1. In Solution Explorer, right-click Data Sources, and then click New Data Source.

2. On the Welcome to the Data Source Wizard page, click Next.

3. On the Select how to define the connection page, click New.

4. In the Server name field, type localhost, in the Select or enter a database name list, click
AdventureWorksDW, click OK, and then click Next.

5. On the Impersonation Information page, click Use a specific Windows user name and password,
enter the user name ADVENTUREWORKS\ServiceAcct and the password Pa55w.rd, and then click
Next.

6. In the Data source name field, type Adventure Works Demo DS, and then click Finish.

Create a Data Source View

1. In Solution Explorer, right-click Data Source Views, and then click New Data Source View.

2. On the Welcome to the Data Source View Wizard page, click Next.
3. On the Select a Data Source page, ensure that Adventure Works Demo DS is selected, and then
click Next.
4. On the Select Tables and Views page, select the DimDate, DimProduct, DimProductCategory,
DimProductSubcategory, DimReseller, DimSalesTerritory and FactResellerSales tables. Add these
tables to Included objects, and then click Next.

5. On the Completing the Wizard page, in the Name field, type Adventure Works Demo DSV, and
then click Finish.
 MCT USE ONLY. STUDENT USE PROHIBITED
2-12 Creating Multidimensional Databases

Modify a Data Source View

1. If it is not already open, in Solution Explorer, right-click Adventure Works Demo DSV, and then click
Open.

2. In Data Source Designer, click the table name of the FactResellerSales table and press F4.

3. In the Properties pane, change the FriendlyName property to Reseller Sales.

4. Repeat the previous steps to rename the DimSalesTerritory table to Sales Territory.

5. Right-click the DimProduct table, and then click New Named Calculation.

6. In the Column name field, type Product Size.

7. In the Expression field, type [Size] + ' ' + [SizeUnitMeasureCode], and then click OK.

8. Right-click the DimProduct table and click Explore Data, then scroll down until you see a value in
the Product Size column (the last column in the table) that shows the size and measurement unit.
9. Close the Explore DimProduct Table window then, on the File menu, click Save All.

10. Keep Visual Studio open. You will return to it in a later demonstration.

Verify the correctness of the statement by placing a mark in the column to the right.

Statement Answer

True or false? You are

investigating data sources for a
planned BI project that includes
Analysis Services
multidimensional cubes. You have
identified an Oracle database that
contains important data, but a
colleague says that you cannot
import data from an Oracle
database into an Analysis Services
cube.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 2-13

Lesson 3
Cubes
A cube is a multidimensional structure containing dimensions and measures. Dimensions define the
structure of the cube and measures provide the numerical values of interest to the end user. As a logical
structure, a cube enables a client application to retrieve values as if cells in the cube define every possible
summarized value. Cubes consume the data that you make available to them through data sources and
data source views.

Lesson Objectives
This lesson describes how to create and modify cubes, and includes some considerations for cube design.
After completing this lesson, you will be able to:

 Create and configure a multidimensional cube.

 Describe the considerations for using time dimensions in a cube.

 Browse a multidimensional cube.

Creating a Cube
To create a cube in Analysis Services, you can use
the Cube Wizard in an Analysis Services project in
SQL Server Data Tools. You can launch the wizard
by right-clicking the Cubes node in a project in
Solution Explorer, and then clicking New Cube.

The Cube Wizard guides you through the process

of creating a cube, and includes various
configuration options:

 You can create a cube by choosing the

relevant data source and data source view,
and then selecting the tables from the view
that you want to include in the cube.
 You can specify the measures from the tables that you want to use.

 When you create the cube, you can add existing dimensions, or create new dimensions that structure
the cube.

 You can also create dimensions separately, by using the Dimension Wizard, and then add them to a
cube.

 Instead of using an existing data source and data source view, you can build the cube without using a
data source and subsequently generate the database schema. Another option is to create the cube,
and enable SSAS to create the underlying schema supporting the cube.
 You can use the Cube Wizard to automatically build attributes and hierarchies, or you can choose to
define them in the Cube Designer later.
 MCT USE ONLY. STUDENT USE PROHIBITED
2-14 Creating Multidimensional Databases

Options for modifying cubes

Although the Cube Wizard creates the cube, many of its properties are available for configuration in the
Cube Designer interface in SQL Server Data Tools. You can use the Cube Designer to:

 View the cube’s properties and its objects.

 Browse cube data.

 Modify cube structure, dimension usage, and calculations.

 Create key performance indicators (KPIs) and actions.

 Configure cube partitioning and storage.

 Create aggregations.

 Create perspectives on cube data.

 Add translations to localize cube data.

Time Dimensions
One of the most important and commonly-used
types of dimension is the time dimension, which
includes time and date values that users can use to
filter the data in the cube. Although most cubes
include a time dimension, some cubes use
different factors instead—such as a unit of work or
project—to delineate data.

Hierarchies in Time Dimensions

Like many dimensions, time dimensions include
hierarchical data; for example, days roll up into
months, and months roll up into years. Depending
on the business requirements of the solution, a
time dimension might include multiple hierarchies. For example, in addition to the days, months, and
years hierarchy, you might include a hierarchy with levels for days, weeks, and years. Together, these two
hierarchies would enable users to explore the data by weeks or by months. However, a single hierarchy
that included years, months, weeks, and days would not work, because weeks do not fit into months in a
regular way. You can create multiple time dimensions, or just a single time dimension that has multiple
hierarchies. In the earlier example, you would typically create one time dimension with multiple
hierarchies but, if you require a time dimension for a different purpose, such as fiscal time—which often
differs from calendar time—you could create it as a separate dimension.

Time Dimension logic in Analysis Services

Analysis Services includes built-in logic for time dimensions that puts the correct days into the correct
months and the correct months into the correct quarter. Time dimensions are the only dimension type
that can include this type of logic, because time data behaves in a consistent way across all databases. For
example, Analysis Services understands how months roll up into years, and can apply this logic in any
database. However, Analysis Services has no way of understanding the logic that controls which product
should be assigned to each product group for any given database, because this will differ between
databases.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 2-15

Types of Time Dimensions in Analysis Services

Creating a time dimension involves creating a table that contains the time data. The dimension is then
created, based on this table. You can use the Dimension Wizard to create a time dimension—this wizard
enables you to create a time dimension in two ways:

 Server time dimensions. Server time dimensions are based on tables stored on the Analysis Services
server. A server time dimension avoids the need for joins in the data source, and can therefore be
faster to process. You can use server time dimensions if security settings prevent you from creating
tables in the data source.

 Time dimension tables. Time dimensions are based on tables that are created and stored in the data
source. Many time dimensions need to include additional data—for example, corporate time
information such as quarters, which do not follow calendar quarters or public holidays. A time
dimension table gives you greater control so you can, for example, add detail to the time members,
which allows you to store additional properties for each date.

Browsing a Cube
After you create a cube, users can access it and
analyze the data that it contains by using a tool
such as Excel. To ensure that the cube functions as
expected before allowing users to access it, you
can use the Cube Browser in SQL Server Data
Tools. You can use the Cube Browser to browse
the cube, and to see the measures that it contains
aggregated by its dimensions. Being able to
browse the cube from within the development
environment is very useful, because it means you
can test the cube without switching between a
client application and SQL Server Data Tools. By
using the Cube Browser, you can change the security context and browse the cube as another user, which
helps to ensure that any security features and perspectives that you have configured are functioning
correctly.

Note: Before you can browse cube data in the Cube Browser tab, you must deploy the
database and cube. You can then click the Browser tab from within the Cube Designer to browse
the cube.

Analyze in Excel
You can also use the Analyze in Excel option, in the Cube Browser in SQL Server Data Tools, to open an
Excel workbook and use it to browse data in the cube. The Analyze in Excel feature automatically creates
the data connection from Excel to the cube, so there is no need to configure it manually.
 MCT USE ONLY. STUDENT USE PROHIBITED
2-16 Creating Multidimensional Databases

Demonstration: Creating and Browsing a Cube

In this demonstration, you will see how to:

 Create a cube.

 Deploy and browse a cube.

 Customize cube measures and dimensions.

Demonstration Steps
Create a Cube

1. Ensure that you have completed the previous demonstration in this module.

2. Maximize Visual Studio, which should be open with the Demo project loaded.

3. In Solution Explorer, right-click Cubes, and click New Cube.

4. On the Welcome to the Cube Wizard page, click Next.

5. On the Select Creation Method page, ensure that Use existing tables is selected, and click Next.

6. On the Select Measure Group Tables page, click Suggest then note that the wizard identifies
DimReseller and Reseller Sales as potential measure group tables. Clear DimReseller so that only
FactReseller Sales is selected, and then click Next.

7. On the Select Measures page, clear the FactReseller Sales check box to clear all selections, select
the Total Product Cost and Sales Amount measures, and then click Next.

8. On the Select New Dimensions page, clear the Reseller Sales dimension, and then click Next.

9. On the Completing the Wizard page, change the cube name to SalesDemo and click Finish. The
cube is created and opened in the Cube Designer.

Deploy and Browse a Cube

1. In Solution Explorer, right-click Demo and click Deploy. If prompted, enter the password Pa55w.rd
for the ADVENTUREWORKS\ServiceAcct user.

2. After the deployment completes, in the Cube Designer, click the Browser tab.
3. In the Metadata pane, expand Measures and expand Reseller Sales. Note that the wizard has
created the measures you selected.

4. Drag the Sales Amount measure to the Drag levels or measures here to add to the query area.
The total sales amount for all reseller sales is shown.

5. In the Metadata pane, note that the wizard has created dimensions for the DimReseller and
DimProduct tables. It has also determined that there are three relationships defined between the
Reseller Sales table and the DimDate table, and so has a dimension for each related column (Due
Date, Order Date, and Ship Date).

6. Expand each dimension and note that the wizard has only created attributes for key columns. You will
need to modify the dimensions to add meaningful attributes for analysis.

7. Drag the Product Key attribute from the Dim Product dimension to the left of the Sales Amount
value in the grid. The browser now shows the aggregated sales amount totals for each product key.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 2-17

Customize Measures and Dimensions

1. In the Cube Designer, click the Cube Structure tab, and then, in the Measures pane, expand Reseller
Sales.

2. Right-click the Total Product Cost measure and click Rename, then change the name of the
measure to Cost.

3. In the Dimensions pane, right-click Dim Reseller, click Rename and change the name of the
dimension to Reseller.

4. Repeat the previous step to rename Dim Product to Product.

5. Right-click the Product dimension and click Edit Dimension. This opens the Dimension Designer for
the Dim Product dimension (the dimension is named Product in the SalesDemo cube, but still
named Dim Product in the database).

6. In the Data Source View pane, in the DimProduct table, drag the EnglishProductName field to the
Attributes pane.

7. In the Attributes pane, right-click English Product Name and click Rename, then rename the
attribute to Product Name.

8. In Solution Explorer, right-click Demo and click Deploy. If prompted, enter the password Pa55w.rd
for the ADVENTUREWORKS\ServiceAcct user.
9. After the deployment completes, click the SalesDemo.cube [Design] tab to return to the Cube
Designer, and then click the Browser tab.
10. On the Browser tab, click the Reconnect button in the upper left corner. If the grid contains any data
from previous queries, right-click in the grid and click Clear Grid.

11. In the Metadata pane, expand Measures and Reseller Sales, and drag Cost to the Drag levels or
measures here to add to the query area.
12. Expand the Product dimension, and then drag Product Name to the left of the Cost value. The total
cost associated with sales for each product is shown.

13. On the File menu, click Save All.

14. Keep Visual Studio open. You will use it again in a later demonstration.
 MCT USE ONLY. STUDENT USE PROHIBITED
2-18 Creating Multidimensional Databases

Check Your Knowledge

Question

You have created a new cube by using

SQL Server Data Tools, and you want to
test it by browsing the cube by using a
PivotTable in Excel. What is the easiest
way to do this?

Select the correct answer.

Open Excel, connect to the cube,

create a PivotTable, and then add
measures and attribute values to it.

In the Cube Browser window in SQL

Server Data Tools, use the Analyze in
Excel options to open an Excel
workbook, and then add measures
and attribute values to it.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 2-19

Lesson 4
Overview of Cube Security
After you install an instance of SQL Server Analysis Services, only members of the server role have server-
wide permissions to perform any task within that instance. This is because, by default, no other users have
access permissions to the objects in the instance. Members of the SSAS server role can grant other users
access to server and database objects by using Microsoft SQL Server Management Studio, SQL Server Data
Tools, or an XML for Analysis (XMLA) script. This lesson describes the SSAS security model, and explains
how to use roles and permissions to implement security.

Lesson Objectives
After completing this lesson, you will be able to:

 Describe the SSAS security model.

 Describe the use of roles in SSAS.

 Describe the use of permissions in SSAS.

The Analysis Services Security Model

SSAS enables users to connect by using only
Windows® accounts─unlike a SQL Server database
engine instance. An Analysis Services instance does
not include the ability to create logins. In a
domain environment, Active Directory user
accounts are authenticated by using the Kerberos
protocol. After successful authentication, users can
access the resources in Analysis Services, providing
that they have been authorized to do so.
Administrators can authorize users by assigning
permissions that allow them access to specific
resources, such as cubes and dimensions. To
manage resource access in Analysis Services, you create roles, assign the required permissions to the roles,
and then assign user accounts to the appropriate roles. User accounts inherit any permissions assigned to
the roles that they belong to.

 SSAS has a single fixed server role for administrators—this is called the server administrator role.
Members of this role have full permissions to perform any action in the instance.

 Administrators can create database roles and grant access permissions to these roles to control user
access and database administrator rights. Role permissions are database specific.

 You can grant database role permissions for Analysis Services databases, cubes, dimensions,
dimension members, individual cells within a cube, mining structures, mining models, data sources,
and stored procedures.
 MCT USE ONLY. STUDENT USE PROHIBITED
2-20 Creating Multidimensional Databases

Working with Roles

Roles are a key part of Analysis Services security.
You can use the built-in server administrator
role and user-created database roles to assign
rights and permissions to users.

The Server Administrator Role

The server administrator role is a fixed role that
has administrative rights over an entire Analysis
Services instance—you cannot change permissions
on this role and you cannot create any additional
server roles. Members of the server
administrator role can access all Analysis Services
databases and objects on that instance. To give
lower-level administrative rights, you should consider adding users to a database role, and granting the
relevant permissions to that role. You can use SQL Server Management Studio to add members to the
server role on the Security page of the Analysis Server Properties dialog box.

Database Roles
To assign permissions to users who do not require server-wide administrative rights, you can create
database roles. You should first decide what levels of access are required within your organization,
including basic users, power users, managers, and department administrators. After you have defined the
roles, you can create them, and then add users to the roles. Consider the following points when planning
database roles and permissions:

 When assigning permissions, you should always follow the principle of least permission—you should
never give a role greater permissions than it requires to enable users to perform their tasks.

 Remove groups with duplicated permissions and, where possible, consolidate groups with similar
permissions, providing this does not give any sensitive permission to another group.

You can create database roles by using SQL Server Management Studio or by using SQL Server Data
Tools:
 In SQL Server Management Studio, in Object Explorer, expand the database, right-click the Roles
folder, and then click New Role.

 In Visual Studio, in Solution Explorer, right-click the Roles folder and then click New Role. You can
also add a role by clicking the Project menu, and then clicking New Role.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 2-21

Assigning Permissions
Administrators assign permissions to roles, and
these permissions determine the actions that a
user can perform. You can set permissions for
Analysis Services objects including databases, data
sources, cubes, cells, dimensions, and dimension
data.

Database Permissions
Assigning permissions to roles in an Analysis
Services database gives users administrative rights
in the database─it does not enable them to view
or work with the data in the database. To enable
users to browse dimensions and measures, you
need to assign permissions on cubes and other objects. The permissions you can assign on Analysis
Services databases include:

 Full control (Administrator). This gives users the ability to perform all administrative tasks, such as
processing databases and database objects, and creating and managing roles.

 Process database. This enables users to process the database. Processing a database populates it
with data from the data source, and updates the database with new data.

 Read definition. You can assign this to roles that also have the Process database permission to
enable users to process a database, by using tools such as SQL Server Data Tools. Without the Read
definition permission, the Process database permission only enables user accounts to process
databases by using scripting.

Data Source Permissions

In most cases, you do not need to assign permissions to enable users to use data sources, because the
permissions that you assign on the cube and other objects are sufficient. However, some data mining
operations, such as performing predictions based on a mining model, do require data source access. In
this situation, you can give users the Read permission on the relevant data source.

Cube Permissions
To access data within the database, users must either be members of a role with permissions to access the
cube, or members of a role with administrative rights. You can assign the following permissions for a cube:

 Read. This enables users to read the data in a cube. This permission is usually sufficient for the
majority of users, who will only need to analyze data and not update it.

 Read/Write. This enables users to read and change the data. Note that updates that users perform
are not written back to the source database—instead, the changes are written to a writeback table.

By default, a role with permissions for the cube can access all objects in the cube, including:

 The individual cells within the cube.

 The dimensions and dimension members that the cube contains.

Note: Permissions to access a data source are not required for a user to access cube data.
 MCT USE ONLY. STUDENT USE PROHIBITED
2-22 Creating Multidimensional Databases

Cell Permissions
Cell permissions control access to measure data in a cube at the individual cell level. A role that has Read
or Read/Write permissions on a cube can, by default, access all individual cells in the cube, because the
permissions are inherited from the cube level. By applying cell permissions, you can override the default
permissions to control access at a more granular level. Although the default is that cell permissions apply
to all cube cells, if you grant access to one or more cells for a role, that role is denied access to all other
cells—until you grant further permissions. You can use an MDX statement to specify the range of cells to
which the permissions apply.

Note: When a user attempts to access cells for which they do not have permission, the
query returns the value #N/A for that cell.

Note that assigning permissions to roles at the cell level cannot raise the permission level that is assigned
to the role at the cube level. For example, if a role has Read permission on a cube and is assigned
Read/Write permission on a cell, the effective permission that they have on the cell is only Read─the
Write portion of the permission is not applied.

In addition to the Read and Read/Write permissions for cells, you can also assign the Read-Contingent
permission. This permission conditionally enables access to a calculated cell, but users can only read the
cell if they also have permissions on all of the cells that the calculation references. For example, imagine a
role called Sales Manager that has the following access rights:

 Read access to the Sales Value measure, but no access to the Product Cost measure.

 Read access to the Profit calculated measure, which is defined as Sales Value minus Product Cost.

When granted Read permission on the Profit measure, Sales Manager role members can see the results
of the Profit measure. Because they can also view the results of the Sales Value measure, they could
calculate the value of the Product Cost measure, even though they have no direct access to it. Applying
Read-Contingent access to the Profit measure means that users can only see the results of the Profit
measure if they also have permission to view both the Sales Value and the Product Cost measures.

Permissions for Dimensions and Dimension Data

Dimension permissions control administrative actions, such as processing the dimension. Dimension
permissions do not control access to the data that a dimension contains. However, you can define
permissions on dimension attribute members and measures to limit access to dimension data.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 2-23

Demonstration: Implementing Cube Security

In this demonstration, you will see how to:

 Create a role and assign permissions.

 Test permissions.

Demonstration Steps
Create a Role

1. Ensure you have completed the previous demonstration in this module. Visual Studio should be open
with the Demo project loaded.

2. In Solution Explorer, right-click Roles, and click New Role. In Solution Explorer, right-click the
Role.role icon that has been added, click Rename, and change the name to Restricted User.role.
When prompted to change the object name, click Yes.

3. In the role designer, on the General tab, note that by default this role has no database permissions.

4. On the Membership tab, note that this is where you can add Windows users and groups to the role.
5. On the Data Sources tab, note that by default the role has no access to any data sources.

6. On the Cubes tab, change the Access value for SalesDemo to Read. This allows the role to access
the cube and read its data.
7. On the Cell Data tab, note that this is where you can specify MDX statements that define cells in the
cube that you want to permit or deny this role to access.
8. On the Dimensions tab, note that the role has Read access to all dimensions in the database. In the
Select Dimension Set list at the top of the page, click SalesDemo cube dimensions and note that
these permissions are inherited by the dimensions in the Demo Cube cube.
9. On the Dimension Data tab, in the Dimension list, under SalesDemo, select Measures Dimension
and click OK. On the Basic tab, clear the check box for the Cost measure.

10. In the Dimension list, under SalesDemo, select Product and click OK. On the Basic tab, in the
Attribute hierarchy list, select Product Name.

11. Select Deselect all members then scroll to the bottom of the list of product names and select all of
the products that begin with the word Touring.

12. On the File menu, click Save All.

Test Permissions

1. In Solution Explorer, right-click Demo and click Deploy. If prompted, enter the password Pa55w.rd
for the ADVENTUREWORKS\ServiceAcct user.
2. When deployment is complete, in Solution Explorer, double-click SalesDemo.cube to return to the
Cube Designer.

3. On the Browser tab, click the Reconnect button at the top left. If the grid contains any data from
previous queries, right-click in the grid and click Clear Grid.

4. In the Metadata pane, expand Measures and Reseller Sales, and drag Sales Amount and Cost to
the Drag levels or measures here to add to the query area. Expand the Product dimension and
drag Product Name to the left of the measure values. The browser shows the cost and sales amount
for all products.

5. Right-click in the grid and click Clear Grid.

 MCT USE ONLY. STUDENT USE PROHIBITED
2-24 Creating Multidimensional Databases

6. In the top left area of the browser window, click the Change User button. In the Security Context –
Demo Cube dialog box, select Roles, and in the drop-down list, select Restricted User and click OK.
Click OK again to return to the browser window.

7. Note that the grid has been cleared then, in the Metadata pane, expand Measures and Reseller
Sales, and drag Sales Amount to the Drag levels or measures here to add to the query area. Note
that members of this role cannot access the Cost measure.

8. Expand the Product dimension and drag Product Name to the left of the measure values. The
browser only shows the sales amount for the Touring products.

9. On the File menu, click Save All, and then close Visual Studio.

Check Your Knowledge

Question

You wish to assign permissions in Analysis Services

to meet the following criteria:
User A needs access to the entire cube to browse
and analyze the data.
User B needs access to the entire cube to browse
and analyze the data, and also needs to perform
predictive analysis against a data mining model.

How should you configure these users’ roles and

permissions?

Select the correct answer.

Create a single role and assign permissions that

enable access to the cube and the data source.
Assign both user accounts to this role.

Create a single role and assign permissions that

enable access to the cube. Assign both user
accounts to this role.

Create a single role and assign permissions that

enable access to the data source. Assign both
user accounts to this role.

Create two roles, named Role X and Role Y.

Assign permissions that enable access to the
cube on Role X and assign permissions that
enable access to the cube and the data source
on Role Y. Assign User A to Role X and User B to
Role Y.

Create two roles, named Role X and Role Y.

Assign permissions that enable access to the
cube on Role X and assign permissions that
enable access to the cube and the data source
on Role Y. Assign User A to Role Y and User B to
Role X.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 2-25

Lesson 5
Configuring SSAS
It is important to understand how SSAS allocates and manages memory to ensure that memory usage and
limits are optimized for the server that is running SSAS. Different configurations, such as multi-instance,
sharing with a SQL Server relational instance, or single-instance configurations, might require some
adjustment of the memory configuration settings. In addition, it is important to understand the optimal
storage configuration for Analysis Services, because this has specific characteristics that are different to
other SQL Server workloads.

Lesson Objectives
After completing this lesson, you will be able to:

 Describe memory configuration settings.

 Describe NUMA support settings.

 Describe common disk configurations.

Configuring SSAS Memory Limits

SSAS allocates an initial amount of memory (40 to
50 MB) when the service starts. This is the case
even if the Analysis Services instance is idle. As
query or processing workloads increase, this
amount of memory is typically increased, based on
certain threshold limits set on the SSAS instance.
The threshold limits are outlined in the following
table:

Setting Description

LowMemoryLimit This limit is used by multidimensional instances, and is the lowest limit at
which the SSAS instance will begin to release memory allocated to
infrequently used objects.

VertiPaqMemoryLimit This limit is used by tabular instances, and is the lowest threshold at
which the SSAS instance begins releasing memory allocated to
infrequently used objects. If memory paging to disk is enabled, this value
also defines the level of memory consumption at which memory paging
should start. The default value is 60.

TotalMemoryLimit This is an upper threshold at which the SSAS instance begins releasing
memory more aggressively to make room for currently executing
requests and new high-priority requests. This limit applies to both
multidimensional and tabular instances. The default value is 80 percent
of either physical memory or virtual address space, whichever is smaller.
 MCT USE ONLY. STUDENT USE PROHIBITED
2-26 Creating Multidimensional Databases

Setting Description

HardMemoryLimit This threshold defines a level at which the SSAS instance starts to reject
requests completely due to memory usage. The default value for this
setting is 0, which automatically places the threshold midway between
the TotalMemoryLimit value and either the physical memory or virtual
address space, whichever is smaller.

Setting the above properties with a value between 1 and 100 will interpret the values as a percentage of
system memory (or virtual address space). Setting the properties with a number larger than 100 will
interpret the value as a fixed number of bytes. For example, setting the value of TotalMemoryLimit to 85
will mean that SSAS will release memory more aggressively when 85 percent of memory is in use, but
setting the value to 12884901888 would mean that SSAS would start releasing memory more aggressively
after 12 GB of memory was in use.

In addition, for tabular instances, you can set the VertiPaqPagingPolicy property to 0 to disable memory
paging to disk. This will generate an out-of-memory error if no more memory is available. Setting the
value to 1 enables paging to disk.

Note: Several other server properties can affect memory behavior, but you should not alter
others than those outlined here without seeking guidance from Microsoft Support.

Typical SSAS Deployments

A few common deployment scenarios are worth considering when deciding whether to adjust the
memory limits values for an Analysis Services instance.

Single Server, Single SSAS Instance

With a single server configuration that is dedicated to a single instance of SSAS, the default memory limits
configuration does not often require any adjustment until the server has 64 GB RAM or more. Higher
memory configurations can leave a large amount of memory reserved for the operating system, which is
unnecessary.

Single Server, Multiple SSAS Instances

SSAS attempts to increase the allocated memory for an instance up to the HardMemoryLimit value, but
does not release this memory back to the operating system or to other applications on the same server.
With multiple SSAS instances sharing the same server, you should configure memory limits carefully so
that one SSAS instance does not restrict memory from other instances. This means that you should take
the value of physical memory in the server, subtract an amount to reserve for the operating system, and
divide the remaining value between all the SSAS instances on the server to set the HardMemoryLimit
value for each instance (typically as a static value in bytes rather than a percentage value). You then set
the TotalMemoryLimit and LowMemoryLimit values for each instance in line with the HardMemoryLimit
for that instance. Note that memory does not need to be divided up evenly between each instance. For
example, you might have a production SSAS instance and a development instance on the same server.
The development instance might have a small value set for HardMemoryLimit, so that there is plenty of
memory available for the production instance which would have a high HardMemoryLimit value.

Single Server, SQL Server DB Instance and SSAS Instance

Although the SQL Server service can grow and shrink memory usage, Analysis Services cannot reduce
memory that has been allocated. This means that, to leave memory available for the SQL Server database
instance, it is important to adjust the memory limit values for Analysis Services. For example, on a server
with 64 GB RAM, you might set the HardMemoryLimit value to 28 GB, to leave the remaining 36 GB for
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 2-27

the operating system and for the SQL Server database instance. In addition, for a configuration such as
this, you might also consider configuring the minimum server memory property on the SQL DB instance,
to ensure that the SQL Server database instance has sufficient memory for processing.

For more information, see Memory Properties in Microsoft Docs:

Memory Properties
https://aka.ms/wc4yue

NUMA Configurations
Non-Uniform Memory Access (NUMA) is a
mechanism designed to improve performance on
multiprocessor systems. NUMA architecture
allocates specific areas of memory to individual
processors in a multiprocessor system, so that
processors experience less contention over access
to a large shared memory.

Analysis Services multidimensional instances are

NUMA node aware. You can use the
PerNumaNode setting to specify NUMA node
behavior for multidimensional instances in the
following ways:

PerNumaNode value Behavior

-1 This is the default setting and enables SSAS to automatically adjust

NUMA node behavior. On systems with less than four NUMA nodes, the
SSAS instance will behave in the same way as setting PerNumaNode=0.
On systems with four or more NUMA nodes, the instance will behave in
the same way as setting PerNumaNode=1.

0 The SSAS instance will ignore NUMA nodes and create one IOProcess
thread pool to be used by all logical processors.

1 The SSAS instance creates one IOProcess thread pool for each NUMA
node. This improves coordinated access to memory.

2 The SSAS instance creates one IOProcess thread pool for each logical
processor. It is intended for highly scaled hardware, with intensive
Analysis Services workloads.

As of SQL Server 2016 SP1 and later, Analysis Services tabular instances are NUMA node aware. NUMA
node awareness and changes to behavior are automatically activated on systems with four or more NUMA
nodes, where a separate job queue is maintained for each NUMA node. On systems with only two NUMA
nodes, the marginal benefit usually doesn’t overcome the overhead of managing NUMA.
For more information, see Improving Analysis Services Performance and Scalability with SQL Server 2016
Service Pack 1 on the Analysis Services Team Blog:

Improving Analysis Services Performance and Scalability with SQL Server 2016 Service Pack 1
https://aka.ms/k6wf45
 MCT USE ONLY. STUDENT USE PROHIBITED
2-28 Creating Multidimensional Databases

Processor Affinity
You can also use the GroupAffinity property to control which Analysis Services thread pools can use which
logical processors in the system. This means a more granular control of workloads for advanced
performance tuning on large, multicore systems. Customizing GroupAffinity on systems that have less
than 64 logical processors might not provide tangible benefits and may even be detrimental to Analysis
Services performance.

For more information, see Thread Pool Properties in the Microsoft Docs:

Thread Pool Properties

https://aka.ms/n3y2ug

Storage Configuration
Based on a typical Analysis Services workload, the
I/O generated is largely random reads during
operational periods. Of course, this depends on
how frequently the cube is being updated, but
cube updates are periodical and often occur
outside of operational hours. This means that disk
configurations that favor read operations are
typically desirable.

In principle, most variants of RAID storage are

suitable from a performance perspective.
However, production environments should also
consider requirements for resiliency. It is
recommended that RAID 1, RAID 5, or another RAID variant that offers some form of redundancy, is used
to protect the cube data from disk error or loss.
In choosing the type of disk, the decision rests on the balance between cost and speed: SATA disks offer
the lowest performance, but are the cheapest; SAS disks offer good speed at a reasonable cost; and SSDs
offer the fastest performance, but at significant cost.

In addition, there are a few general recommendations for Analysis Services storage configuration:

 Configure a single volume. It is not normally considered a good practice to separate any of the data
files for the cube onto a separate volume. Configuration of a single, large volume for your cubes will
typically ease management overhead and allow the largest I/O bandwidth to the storage hardware.

 Exclude Analysis Services folders from virus scanners. Ensure that the Data, TempDir, and backup
folders are not being scanned by antivirus software. There are no executable files in these folders.

 Defragment the data folder. Over time, the files in cube data folders can become very fragmented.
Periodically, you should defragment the cube volume during a maintenance period to keep
performance at expected levels.

For more information, see the Microsoft SQL Server Analysis Services Multidimensional Performance and
Operations Guide on the Microsoft Download Center.

Microsoft SQL Server Analysis Services Multidimensional Performance and Operations Guide
https://aka.ms/q0gr28
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 2-29

Check Your Knowledge

Question

You are configuring an Analysis Services

instance to coexist on the same server as
a SQL Server relational database instance.
You want to ensure that Analysis Services
leaves enough memory for the relational
instance to be always running. What
setting should you configure?

Select the correct answer.

Set the HardMemoryLimit value to 0.

Set the TotalMemoryLimit value to 0.

Set the HardMemoryLimit value to 95.

Set the HardMemoryLimit value to 45.

Set the LowMemoryLimit value to 45.

 MCT USE ONLY. STUDENT USE PROHIBITED
2-30 Creating Multidimensional Databases

Lesson 6
Monitoring SSAS
Monitoring an Analysis Services multidimensional instance for performance will give you important
information regarding instance and cube performance. Of course, instance performance will be affected
by the availability of hardware resources, such as processor or memory. However, cube performance can
also be greatly affected by the way queries are written and processed.

Lesson Objectives
After completing this lesson, you will be able to:

 Select Analysis Services instance counters to monitor.

 Use tools to monitor Analysis Services.

 Select counters to monitor cube and query performance.

Considerations for Analysis Services Performance

Before attempting to identify and analyze the
performance of an Analysis Services instance, it is
important to understand the workloads and the
available tools.

Analysis Services Workloads

Analysis Services databases do not have the same
workload characteristics as relational database
instances, and must typically support the following
two key workloads:

 Queries. Users submit queries to retrieve data

from Analysis Services (typically through
various applications). Both multidimensional and tabular instances must perform two tasks:

o Data retrieval. The storage engine extracts the required data from the data model.

o Calculations. The formula engine performs the required calculations to output the data.
 Cube Processing. The data held in the cube is typically updated at a predetermined interval, to allow
queries to represent more up-to-date information or changes to the data model. Cube processing
can be a resource intensive task, so it should only be performed as frequently as the data needs
updating for fresh reports or analysis output.

Monitoring Analysis Services

Performance monitor has a large number of counters available for monitoring Analysis Services that can
be used for real-time monitoring and troubleshooting, or for creating performance baselines for future
comparison.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 2-31

Analysis Services Performance Counters

Each Analysis Services instance on the server has a dedicated set of performance counters. These and the
system wide counters can help to provide an overview of Analysis Services resource utilization:

Scope Object Counter(s) What the counter shows

Analysis Services Memory Memory Usage KB The quantity of memory

used by Analysis Services.

Analysis Services Connection Current Connections The number of open

connections to Analysis
Services.

Analysis Services Proc Temp file bytes written The amount of data written
Aggregations to the temp file during
processing operations. If the
cube and hardware are well
balanced, this will be near 0.

Analysis Services Proc Indexes Current Partitions Measures the speed and
Rows/sec concurrency of index
processing.

Analysis Services Processing Rows read/sec The speed of processing

Rows written/sec data.

System wide Memory Available Kbytes The amount of physical

Page Faults/sec memory available to the
server and the frequency of
memory paging to disk.

System wide System File Read Bytes/sec File system read activity,
File Read Operations/sec including reads from the file
system cache.

System wide Network Bytes Received/sec Quantity of network traffic.

Interface Bytes Sent/sec

System wide TCPv4 and Segments/sec Stability of network

TCPv6 Segments connections.
Retransmitted/sec

For performance troubleshooting, you might also include Analysis Services counters for Threads, MDX and
Storage Engine Query objects. For I/O troubleshooting, the Memory and Physical Disk objects contain
some useful extra counters.
 MCT USE ONLY. STUDENT USE PROHIBITED
2-32 Creating Multidimensional Databases

Analysis Services Monitoring Tools

SQL Server Profiler enables you to create, capture
and replay event data (called a trace) to examine
behavioral and performance characteristics of
Analysis Services. For example, you might perform
a trace to determine how long certain actions
performed by Analysis Services last, when a
specific query is made.

SQL Server Profiler works by capturing event

information that occurs while Analysis Services is
running. Events are certain types of action, such as:

 Login connections or disconnections.

 Transact-SQL statements.
 The start or end of a stored procedure.

 The start or end of a SQL batch.

 An error written to the log.

These events are grouped into classes. To capture only the information that you need, you can select
which classes of event to capture in the trace. Examples of event classes for Analysis Services include:

 Audit Login

 Audit Logout

 Query Begin

 Query End
 Query Subcube

SQL Server Profiler also gives you the ability to select how much information about each kind of event
you capture. Examples of these DataColumns include:

 ApplicationName

 StartTime

 Duration

 DatabaseName

 NYUserName

When you create a trace for SQL Server Profiler, and select the event classes and DataColumns for the
trace, you can also save the trace as a template. The template is saved as a file with a .tdf extension. This
enables you to rerun the trace at a later time to capture the same kind of information, or transfer the file
to a different server to run the same trace there—such as from a development server to a production
server.

For more information, see SQL Server Profiler in the Microsoft Docs:

SQL Server Profiler

https://aka.ms/mab11d
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 2-33

Note: Note that SQL Server Profiler has been deprecated for use with the SQL Server
database engine. SQL Server Profiler is not deprecated for use with Analysis Services in SQL Server
2017.

Extended Events
Analysis Services also includes a graphical user interface to create and manage Extended Events
(sometimes called XEvents) within SQL Server for capturing detailed event data. Extended Events can
capture information similar to SQL Server Profiler; however, Extended Events are better because they can
capture more information and perform additional tasks—and they have less overhead on the system. For
example, it is reported that 20,000 events/sec on a 2ghz CPU with 1 GB of RAM takes less than 2 percent
of the CPU. This means that, on a modern server, there should be little impact on performance.

Extended Events are constructed in SQL Server Management Studio (SSMS), in a similar way to SQL Server
Profiler traces, starting by selecting the types of event to capture (these include equivalent SQL Server
Profiler events), and the associated information about the event, called event fields. Extended Event data
can be saved to a file for later analysis, or sent to a memory buffer for real-time monitoring. You can also
use SSMS to output your Extended Event configuration as a script, which can be useful to reproduce the
Extended Event easily on other servers, or to reproduce the capture data after you delete the Extended
Event. Deleting the session object is currently the only way to stop the Extended Event from capturing
data.

For more information, see Monitor Analysis Services with SQL Server Extended Events in Microsoft Docs:
Monitor Analysis Services with SQL Server Extended Events
https://aka.ms/M4gpz3

Dynamic Management Views

Analysis Services also provides Dynamic Management Views (DMVs) to review current server state
information. DMVs operate as a set of queries designed to return metadata and monitoring information
about the Analysis Services instance.

Most DMV queries are executed by using a SELECT statement on the $System schema with an XML/A
schema rowset. These include:
 $system.discover_commands that lists all currently running commands on the server.

 $system.discover_connections that lists current open connections.

 $system.discover_sessions that lists all sessions on the server. You can use it to determine the
commands that were executed and in which connections.

 $system.discover_memoryusage that lists all memory allocations in the server.

 $system.discover_locks that lists currently held and requested locks.

For more information, see Use Dynamic Management Views (DMVs) to Monitor Analysis Services in the
Microsoft Docs:

Use Dynamic Management Views (DMVs) to Monitor Analysis Services

https://aka.ms/vx9kt2
 MCT USE ONLY. STUDENT USE PROHIBITED
2-34 Creating Multidimensional Databases

Monitoring Query Performance

In addition to examining server resource issues, it
is also important to recognize that query
construction can affect query performance.
Identifying query performance issues can be
challenging, but monitoring the following
counters may help identify elements that can be
improved:

Counter What to look for

MDX: Total cells calculated If this number is high, queries might be performing cell-
by-cell calculations instead of block-oriented calculations.

MDX: Number of calculation covers The total number of active and cached evaluation nodes
built by MDX execution plans. If this number is high,
queries might be performing cell-by-cell calculations
instead of block-oriented calculations.

MDX: Total Sonar subcubes The total number of subcubes that query optimizer
generated. If this number is high, queries might be
performing cell-by-cell calculations instead of block-
oriented calculations.

MDX: Total recomputes The number of cells recomputed due to error. Non-zero
values indicate recalculations to fix errors, again possibly
on a cell-by-cell basis, which are using additional
processor time.

MDX: Total NON EMPTY The number of times an unoptimized NON EMPTY
unoptimized algorithm is used.

MDX: Total NON EMPTY for The total number of times a NON EMPTY algorithm was
calculated members looping over calculated members.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 2-35

Lab: Creating a Multidimensional Database

Scenario
Business analysts at Adventure Works Cycles have requested an analytical solution that enables them to
“slice and dice” business data to analyze Internet sales. To accomplish this, you will create and test a
multidimensional database that includes an OLAP cube.

Objectives
After completing this lab, you will have:

 Created a data source.

 Created and modified a data source view.

 Created and modified a cube.

 Added a dimension to a cube.

Estimated Time: 45 minutes

Virtual machine: 20768C-MIA-SQL

User name: ADVENTUREWORKS\Student

Password: Pa55w.rd

Exercise 1: Creating a Data Source

Scenario
Information workers want to analyze Internet sales data in Excel. To support this, you plan to implement a
multidimensional database. First, you need to create a connection and data source to the data warehouse.

The main tasks for this exercise are as follows:

1. Prepare the Lab Environment

2. Create an Analysis Services Project

3. Create a Data Source

 Task 1: Prepare the Lab Environment

1. Ensure that the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and then
log on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd.

2. Run Setup.cmd in the D:\Labfiles\Lab02\Starter folder as Administrator. Note that the script may
take up to several minutes to complete.

 Task 2: Create an Analysis Services Project

1. Use SQL Server Data Tools to create a new Analysis Services Multidimensional and Data Mining
project.
2. Name the project Adventure Works OLAP, and then save it in the D:\Labfiles\Lab02\Starter
folder.

 Task 3: Create a Data Source

1. Use the Data Source Wizard in Visual Studio to create a data source for the AdventureWorksDW
database on the local server.
2. Use Windows authentication to connect to the data source.
 MCT USE ONLY. STUDENT USE PROHIBITED
2-36 Creating Multidimensional Databases

3. Configure the data source to use the ADVENTUREWORKS\ServiceAcct account with the password
Pa55w.rd.
4. Name the data source Adventure Works Data Warehouse.

Results: After completing this exercise, you will have:

Created an Analysis Services project by using SQL Server Data Tools.

Created a data source.

Exercise 2: Creating and Configuring a Data Source View

Scenario
The AdventureWorksDW database includes a large amount of data not required by your users, and the
tables use naming conventions that are confusing to business analysts. You need to define which tables
you require, and then change their names to be more user-friendly. Users have also requested that they
can retrieve full names of customers, in addition to their first, middle, and last names.

The main tasks for this exercise are as follows:

1. Create a Data Source View

2. Configure a Data Source View

 Task 1: Create a Data Source View

1. Use the Data Source View Wizard in Visual Studio to create a data source view that references the
Adventure Works Data Warehouse data source that you created in the previous task.

2. Include the following tables:

o DimCustomer
o DimDate

o DimGeography

o DimProduct
o DimProductCategory

o DimProductSubCategory

o FactInternetSales

3. Name the data source view Adventure Works DSV.

 Task 2: Configure a Data Source View

1. Change the FriendlyName property of the tables in the Adventure Works DSV data source view to
remove the “Dim” or “Fact” prefix and add spaces between words where appropriate.

o For example, the FactInternetSales table should have the friendly name Internet Sales, and the
DimProductCategory table should have the friendly name Product Category.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 2-37

2. Add a calculation named Full Name to the Customer table. Use the following MDX expression:

CASE
WHEN MiddleName IS NULL THEN
FirstName + ' ' + LastName
ELSE
FirstName + ' ' + MiddleName + ' ' + LastName
END

Results: After completing this exercise, you will have:

Created a data source view.

Configured a data source view.

Exercise 3: Creating and Configuring a Cube

Scenario
Now you have created a data source view, you are ready to create a cube that enables business users to
perform multidimensional analysis on the data. You will then modify the cube to give the measures user-
friendly names to make analysis simpler. You will also add extra attributes to dimensions, to provide users
with multiple ways to aggregate the measures.
You need to modify dimension attributes used for dates in your cube to ensure uniqueness across
temporal periods. For example, the month of January is not unique because it occurs in every year. You
must modify the month attribute so that its key is based on both month and year, and then specify which
of these key columns should be used when displaying the attribute name.

The main tasks for this exercise are as follows:

1. Create a Cube

2. Configure Measures
3. Configure Dimensions

4. Browse the Cube

 Task 1: Create a Cube

1. Use the Cube Wizard in SQL Server Data Tools to create a new cube from existing tables.
2. Use the Internet Sales table as the measure group, and select the following measures:

o Order Quantity

o Total Product Cost

o Sales Amount

o Internet Sales Count

3. Create dimensions for all tables other than Internet Sales.

4. Name the cube Sales.

 Task 2: Configure Measures

1. Change the name of the Order Quantity measure to Internet Order Quantity.

2. Change the name of the Total Product Cost measure to Internet Cost.
 MCT USE ONLY. STUDENT USE PROHIBITED
2-38 Creating Multidimensional Databases

3. Change the name of the Sales Amount measure to Internet Revenue.

4. Save the cube.

 Task 3: Configure Dimensions

1. Modify the Customer dimension by using Dimension Designer.

2. Add the City, StateProvinceName, EnglishCountryRegionName, and PostalCode attributes from

the Geography table to the Geography dimension.

3. Add the CustomerAlternateKey, Title, FirstName, MiddleName, LastName, and Full Name
attributes from the Customer table to the Customer dimension.

4. Modify the Product dimension by using Dimension Designer.

5. Add the ProductAlternateKey, EnglishProductName, and ListPrice attributes from the Product
table to the dimension.
6. Add the EnglishProductSubcategoryName attribute from the ProductSubcategory table to the
dimension.

7. Add the EnglishProductCategoryName attribute from the ProductCategory table to the

dimension.
8. Modify the Date dimension by using Dimension Designer.

9. Add the EnglishMonthName, MonthNumberOfYear, CalendarQuarter, CalendarYear, and

CalendarSemester attributes from the Date table to the dimension.
10. Rename the English Month Name attribute to Month.

 Task 4: Browse the Cube

1. Deploy the Adventure Works OLAP project, entering the user name
ADVENTUREWORKS\ServiceAcct and the password Pa55w.rd if prompted.

2. Use the Cube Browser in Visual Studio to view the Sales cube, and view the Internet Revenue and
Internet Sales Count measures to the data area by Order Date.Calendar Year.

Results: After completing this exercise, you will have:

Created a cube.

Edited measures in the cube.

Edited dimensions in the cube.

Browsed the cube.

 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 2-39

Exercise 4: Adding a Dimension to the Cube

Scenario
After you have created the cube with the wizard, a business analyst has requested the ability to view
Internet sales by sales territory. To accommodate this, you must create a dimension in the database, and
add it to the cube.

The main tasks for this exercise are as follows:

1. Edit the Data Source View

2. Create a Dimension

3. Add a Dimension to a Cube

4. Analyze a Cube by Using Excel

 Task 1: Edit the Data Source View

1. Edit the Adventure Works DSV data source view and add the DimSalesTerritory table.

2. Change the friendly name of the DimSalesTerritory table to Sales Territory.

3. Save the data source view.

 Task 2: Create a Dimension

1. Add a dimension to the database based on the Sales Territory table you added in the previous task.

2. Use SalesTerritoryKey as the key column and the name column.

3. Add the SalesTerritoryRegion, SalesTerritoryCountry, and SalesTerritoryGroup attributes and
enable browsing for each of them.

 Task 3: Add a Dimension to a Cube

1. Add the Sales Territory dimension to the Sales cube.
2. Verify that the Sales Territory dimension is used by the Internet Sales measure group through a
relationship based on the Sales Territory Key attribute.

3. Save the cube.

 Task 4: Analyze a Cube by Using Excel

1. Deploy the Adventure Works OLAP project, entering the user name
ADVENTUREWORKS\ServiceAcct and the password Pa55w.rd if prompted.

2. Analyze the cube in Excel, and view the Internet Revenue measure by the Sales Territory Group,
Sales Territory Country, and Sales Territory Region attributes of the Sales Territory dimension.

3. Close Excel without saving the workbook.

4. Save the project and close SQL Server Data Tools.

Results: After completing this exercise, you will have:

Edited the data source view.

Created a dimension.
Added a dimension to the cube.

Browsed the cube by using Excel.

 MCT USE ONLY. STUDENT USE PROHIBITED
2-40 Creating Multidimensional Databases

Question: In the final exercise, in which you added a dimension to the cube, you edited the
data source view to include the table that the dimension was based on. Why was it not
necessary to also edit the data source object?
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 2-41

Module Review and Takeaways

In this module, you learned about:

 The core components of a multidimensional database solution.

 Data sources and data source views.

 Creating and browsing multidimensional cubes.

 How to implement security in a multidimensional database.

Review Question(s)
Question: What should you consider when assigning friendly names for objects in a
multidimensional database?
MCT USE ONLY. STUDENT USE PROHIBITED
 
 MCT USE ONLY. STUDENT USE PROHIBITED
3-1

Module 3
Working with Cubes and Dimensions
Contents:
Module Overview 3-1 
Lesson 1: Configuring Dimensions 3-2 

Lesson 2: Defining Attribute Hierarchies 3-7 

Lesson 3: Implementing Sorting and Grouping for Attributes 3-13 

Lesson 4: Slowly Changing Dimensions 3-16 

Lab: Working with Cubes and Dimensions 3-19 

Module Review and Takeaways 3-24 

Module Overview
Dimensions are a fundamental component of cubes. Facts, or measures, have the values that interest you,
but dimensions provide the business context by which you aggregate these measures. Dimensions
organize your data by category and show results grouped by these categories, such as product, customer,
or month.

Microsoft® SQL Server® Analysis Services dimensions contain attributes that correspond to columns in
dimension tables. These attributes appear as attribute hierarchies and can be organized into user-defined
hierarchies, or defined as parent-child hierarchies, based on columns in the underlying dimension table.
You use hierarchies to organize measures contained in a cube.

Objectives
After completing this module, you will be able to:
 Configure dimensions.

 Define attribute hierarchies.

 Sort and group attributes.

 Configure Slowly Changing Dimensions (SCDs).
 MCT USE ONLY. STUDENT USE PROHIBITED
3-2 Working with Cubes and Dimensions

Lesson 1
Configuring Dimensions
All Analysis Services dimensions are groups of attributes based on columns from tables or views in a data
source view. Dimensions can exist independently of a cube—you can use dimensions in multiple cubes,
use them multiple times in a single cube, and link them between Analysis Services instances. A database
dimension exists independently of a cube; an instance of a database dimension within a cube is called a
cube dimension.

Lesson Objectives
After completing this lesson, you will be able to:

 Describe the concept of a dimension.

 Edit dimensions with Dimension Designer.

 Describe the dimension storage options.

 Describe the use of dimension attributes.

 Define the columns used to connect dimensions to the fact table, to display to users, and to use in
Multidimensional Expressions (MDX) calculations.

Dimension Concepts
Dimensions form the contexts for facts, and define
the aspects of the business by which you
aggregate facts. For example, Product and Time
dimensions might provide context to the Sales
Revenue measure.

Dimensions are collections of attributes from

tables or views. You can use dimensions in
multiple cubes and link to them from remote
instances.
Attributes add meaning to dimensions. Each
column in the dimension table can provide an
attribute that has a piece of information about the
dimension member. For example, a Retail Store dimension might have Manager and Postal Code
attributes.

The Key attribute is typically the primary key of the dimension table. The Key attribute is the column in
the dimension table that links to the fact table. The Name attribute provides a friendly name for the
dimension, enabling you to display an alternative, more relevant, name than the source table name-to-
end users.
Typically, you arrange attributes into hierarchies that define the drill-down paths through aggregations.
Each layer of the hierarchy is called a level. For example, a Time dimension might have a hierarchy with
Day, Month, and Year levels, but you could define other hierarchies that have attributes such as Week,
Quarter, and Century.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 3-3

By default, every attribute defined within a dimension is part of a two-level hierarchy known as an
attribute hierarchy. The top level is named All; the second is the Leaf level that contains values for each
individual member of the attribute. While they can help in browsing, the data within an attribute
hierarchy can typically be too large and difficult to navigate. Attribute hierarchies are often replaced with
user defined hierarchies where you can define more levels and make data more manageable. For example,
a Date attribute hierarchy might have an All level and a second level containing all the dates within the
dimension. To make this more manageable, you could replace the Date attribute hierarchy with a user-
defined hierarchy named Calendar Year that contains Year, Quarter and Month levels, making the data
easier to navigate.

Analysis Services can add dimension intelligence to some dimensions. For example, time dimensions have
a fixed structure that can be used by Analysis Services. It knows which months are in each quarter, which
days are in each month, and so on.

The Dimension Designer

Use the Dimension Designer in SQL Server Data
Tools to edit the attributes, levels, hierarchies, and
translations of a dimension, and to browse the
dimension. The Dimension Designer includes four
tabs:

 Use the Dimension Structure tab to view and

edit the attributes, levels, and hierarchies of
the dimension.

 Use the Attribute Relationships tab to

create, modify, or delete the attribute
relationships of the selected dimension.

 Use the Translations tab to view and edit the multilanguage translations for the attributes, levels,
and hierarchies.

 Use the Browser tab to browse members of each hierarchy in the dimension. You can only browse
members after you have deployed the solution.

Configuring Dimension Storage

The two dimension storage modes available in
Analysis Services are Multidimensional Online
Analytical Processing (MOLAP) and Relational
Online Analytical Processing (ROLAP). These
storage modes define where, and in what structure
type, the dimension data is stored:
 Data for a dimension that uses MOLAP is
stored in a multidimensional structure in the
instance of Analysis Services. This
multidimensional structure is created and
populated when the dimension is processed.
MOLAP dimensions provide better query
performance than ROLAP dimensions.
 MCT USE ONLY. STUDENT USE PROHIBITED
3-4 Working with Cubes and Dimensions

 Data for a dimension that uses ROLAP is stored in tables used to define the dimension. The ROLAP
storage mode can be used to support large dimensions without duplicating large amounts of data,
but at the expense of query performance. Because the dimension relies directly on the tables in the
data source view used to define the dimension, the ROLAP storage mode also supports real-time
ROLAP.

Configuring Dimension Attributes

The Cube Wizard and Dimension Wizard create
attributes for a dimension. Other wizards and
elements of the Analysis Services user interface
may further modify these attributes. These default
settings are sufficient in most situations, but you
can use Dimension Designer to edit the attributes.
You can remove attributes from the dimension by
right-clicking the attribute and clicking Delete.
Removing attributes from a dimension does not
affect the data source view, which can be used by
multiple dimensions without forcing each
dimension to use all the same attributes.

You can rename an attribute to provide a more meaningful or user-friendly name than the dimension
table. Do this by right-clicking the attribute in the Dimension Structure tab of the Dimension Designer.
You can also set the Name property of the attribute in the Properties window or edit it directly if the
Attributes pane is in the grid view.
Having many attributes for a dimension can confuse users. You can organize attributes into display folders
to simplify browsing. Using folders only affects the way client applications display the dimension and has
no other effects on hierarchies or attributes. After you have created display folders, deploy the solution
and reconnect to see the results on the Browser tab.

Attributes can be used to create hierarchies. For example, a Product dimension might have Product
Category, Product Sub-Category, and Product Name attributes. Hierarchies are covered in more detail
in the next lesson.

Attributes are used to add extra detail to dimension members and do not need to be the basis for a
hierarchy method. Therefore, in the previous example, you might have Product Description and Product
Image attributes. These attributes are not used for a hierarchy, nor to sort or group members, but are
often needed to provide additional detail to dimension members.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 3-5

Attribute Column Bindings

To control output from attributes, you can define
the column that uniquely identifies attribute
values, the column that users see, and an optional
value column you can use for MDX calculations.

 The KeyColumn is the column, or columns,

that uniquely identify members for each
attribute value in the fact table. This is often
the primary key of the dimension table, and is
used when you choose to order the hierarchy
by key. In some cases, a member must be
identified by a combination of column values.
For example, in a geography dimension, the
City value “Paris” might not be unique, as there might be a row for Paris, France and another row for
Paris, Texas, in the United States. You can either use the primary key column of the dimension table
to identify the correct “Paris”, or a multicolumn key that includes the city, state, and country or region
to identify members uniquely in your dimension. This is particularly important if you plan to include
an attribute in a hierarchy.
 The NameColumn provides the value that a user will see and can give more useful information than
the key column. You can define a name column as a calculated column you have created in a data
source view. For example, an Employee dimension might have an EmployeeID key column, but you
might want the name of the employee to be displayed. To do this, you can create a calculated
column based on first and last name in the data source view, and then use this as the name column.

 The ValueColumn is returned by the MDX MemberValue function. This means you can create
calculations based on a value other than the name or key. For example, a time dimension is stored in
a dimension table with one row for each day. The key column is a SMALLINT and the name column is
a CHAR(11) that displays the day, month, and year. The value column contains the date as a
SMALLDATETIME. This allows calculations to use the true date value rather than converting it from
string values, but also enables the key column to hold a smaller value. A key column value occurs in
every row of the fact table, so reducing its size makes the fact table smaller and makes joins between
the fact table and the time dimension faster.
 MCT USE ONLY. STUDENT USE PROHIBITED
3-6 Working with Cubes and Dimensions

Check Your Knowledge

Question

MOLAP dimensions provide better

performance than ROLAP but which of
the following is a reason for choosing
ROLAP?

Select the correct answer.

ROLAP is easier to implement than

MOLAP.

When real-time OLAP is required.

Querying ROLAP generally involves

simpler queries.

ROLAP is more efficient with small

dimensions.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 3-7

Lesson 2
Defining Attribute Hierarchies
Hierarchies define the multilevel structure of a dimension, and the relationships between the levels can
affect query and processing performance.

Lesson Objectives
After completing this lesson, you will be able to:

 Describe hierarchies.

 Distinguish parent-child from balanced hierarchies and describe when each type should be used.

 Describe ragged hierarchies and the options you have when configuring them.
 Describe considerations for using hierarchies.

 Create attribute relationships.

What Are Hierarchies?

Attributes are exposed to users through attribute
hierarchies. An attribute hierarchy in a dimension
includes an optional All level and distinct
members of the attribute. For example, a
Customer dimension might include a Name
attribute hierarchy with two levels: the All level
and a second level with a member for each name.
Attribute hierarchies define the space of a cube,
which you can think of as the multidimensional
space created by the product of its attribute
hierarchies. In addition to being containers for
attribute hierarchies, dimensions can also include
user hierarchies as a navigational convenience, but these do not affect cube space.

Defining relationships between hierarchy levels enables Analysis Services to establish more useful
aggregations that, in turn, increase query performance. This can save memory during processing and
increase performance—which is often important with large or complex cubes.
When arranging attributes into user-defined hierarchies, you define relationships between hierarchy
levels. The levels are connected in a many-to-one or one-to-one relationship, referred to as a natural
relationship. For example, in a Calendar Time hierarchy, a Day level is related to the Month level, the
Month level to the Quarter level, and so on.

A natural hierarchy is composed of attributes where each is a member property of the attribute below. For
example, a Customer Geography hierarchy with the levels Country-Region, State-Province, City, ZIP
Code and Customer, is a natural hierarchy because of the relationships between the attributes. By
contrast, a Marital Status-Gender hierarchy containing the levels Marital Status and Gender is non-
natural, because marital status and gender do not have a hierarchical relationship to each other.
For performance reasons, natural hierarchies are preferred and the Dimension Designer will warn you if
you create a hierarchy that is non-natural. You can ignore this warning when it makes sense to use a non-
natural hierarchy. For example, it is plausible that you would want to drill down from marital status to
gender.
 MCT USE ONLY. STUDENT USE PROHIBITED
3-8 Working with Cubes and Dimensions

Parent-Child Hierarchies
A parent-child hierarchy exists when you have a
self-referencing dimension table. For example, an
Employee dimension might have a Manager
attribute. With the exception of the most senior
employee, all other employees have a position in
the hierarchy defined by the identities of their
managers, and those they manage.

A parent-child hierarchy is an unbalanced

hierarchy. This imbalance occurs when the number
of levels from leaf to root is different for individual
leaf members. For example, some managers might
have three levels of employee working for them,
while others have just two.

You define parent-child hierarchies, not by the order in which attributes are added, but by the
relationships in the dimension table.
Parent-child dimensions are “changing dimensions”, which means that rows can be modified without the
need for processing. They are also the only dimension type you can write-enable.
Parent-child dimensions can be read-only and are classified as changing dimensions because the
hierarchy structure is calculated at run time. This can be beneficial if the dimension records change
frequently, but there are performance considerations. Because the parent-child hierarchy is calculated at
run time, no aggregates are calculated when the cube is processed. For example, the total sales for each
sales manager, and every employee who reports to them, are only calculated when the query is run.

This trade-off between a versatile changing dimension and query performance must be considered and,
ideally, the effects tested. If you do not want to use a parent-child dimension, you can create traditional
dimensions. For example, an employee could list their level managers up through the hierarchy. This
would allow the dimension aggregates to be calculated during processing, but with a loss of flexibility.
You should also consider what to do with non-leaf data. In the example used here, the sales manager can
also make sales. If this is the case, and you have data in the fact table associated to non-leaf members,
you must change the dimension's Members With Data property to non-leaf data visible or non-leaf data
hidden. Otherwise, processing the cube fails. If you use non-leaf data visible, the parent—in this case the
sales manager—will also appear in the level below and seem to report to himself or herself. If you use
non-leaf data hidden, the sales manager will not appear in the level below and, should you perform a sum
on all employee sales, the value will not include the sales manager.
For more information on parent-child dimensions, see Parent-Child Dimension in Microsoft Docs:

Parent-Child Dimension
http://aka.ms/Whhuwv
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 3-9

Ragged Hierarchies
To the user, ragged and parent-child hierarchies
often seem similar. They both have a different
number of levels in separate parts of the hierarchy,
but these are formed from different dimension
table columns, rather than from a self-referencing
relationship. For example, in a Location dimension,
you might have Location, Region, State and
Country levels, but the State level might only be
used for the United States, Canada, and Australia.
All other countries might skip the State level.

The HideMemberIf property makes a regular

hierarchy ragged. There are five possible values for
this property:

 Never – this value creates a regular hierarchy.

 OnlyChildWithNoName – this value hides a level member when it is an only child and is null or a
zero-length string.

 OnlyChildWithParentName – this value hides a level member when it is an only child with the same
name as its parent.
 NoName – this value hides a level member when it is null or a zero-length string.

 ParentName – this value hides a level member when it has the same name as its parent.
You can use a ragged hierarchy as an alternative to a parent-child hierarchy. The aggregates are
calculated when the cube is processed, improving query performance. A ragged hierarchy is not, however,
a changing dimension, so the flexibility of parent-child dimensions is lost.

Using Hierarchies
You typically create user-defined hierarchies to
allow users to drill up and drill down through the
data, and to allow Analysis Services to create
meaningful aggregations of the measures.

There are several important hierarchy properties:

 The IsAggregatable property defines

whether an All level is created. In some
scenarios, an All level is irrelevant. For
example, the Time dimension in a solution
holds data from January 12, 1995, because
earlier data is too costly to convert and gives
minimal business benefit. This range of dates
is not relevant to the business and does not need to be aggregated.
 The AttributeHierarchyOrdered property defines whether the hierarchy is ordered. If this is set to
False and you do not query the attribute hierarchy, you will reduce processing time. If you only use
an attribute hierarchy to order another attribute hierarchy, it is not queried.
 MCT USE ONLY. STUDENT USE PROHIBITED
3-10 Working with Cubes and Dimensions

 The AttributeHierarchyOptimizedState can be set to NotOptimized to prevent Analysis Services

from creating indexes on the attribute hierarchy. This will reduce processing time but increase query
time, and should only be used when you do not query the attribute hierarchy.

 Hiding and disabling attribute hierarchies can improve performance and focus data for users. For
example:
 If you set the AttributeHierarchyEnabled property to False, the hierarchy is disabled and Analysis
Services only creates the attribute hierarchy as a member property. This is useful if the attribute is
providing detail, but you do not want to use it as a level for aggregation.

 If you set the AttributeHierarchyVisible property to False, the attribute is only visible from user-
defined hierarchies. This is useful if there are large numbers of distinct values in the attribute
hierarchy that would cause confusion and not add any benefit.

Attribute Relationships
Attributes are always related directly or indirectly
to the key attribute. Initially all attributes are
directly related to the key attribute and this might
be the relationship you require. For example, a
Product dimension contains ProductKey, Name,
Weight, ProductionCost, and
RecommendedSalesCost. In this dimension, you
do not group by any attribute, and you only need
the All level and one other level containing every
member.
In most dimensions, there are levels with
relationships between them. For example, a
Customer dimension containing attributes for CustomerKey, Name, Address, City, Region, and
Country. Name and Address attributes relate directly to the CustomerKey, but relationships have to be
created for the other attributes. You form a relationship, from Country to Region, from Region to City,
and from City to CustomerKey; this supports levels within the data and is used to create aggregates to
improve query performance.

Attribute relationships are straightforward to create, using the Attribute Relationships tab of the
Dimension Designer. You can right-click a blank space to create a dimension relationship, and then
choose the source and related attributes. Each source attribute can only have one related attribute. For
example, a month can only have one quarter; although a quarter can have several months. In this
relationship, the month is the source attribute and the quarter is the related attribute.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 3-11

Demonstration: Creating a Parent-Child Hierarchy

In this demonstration, you will see how to:

 Create a parent-child hierarchy.

 Name the levels in a parent-child hierarchy.

Demonstration Steps
Create a Parent-Child Hierarchy

1. Ensure that the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and log
on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd. In the
D:\Demofiles\Mod03 folder, run Setup.cmd as Administrator.

2. Start Visual Studio and open Demo.sln in the D:\Demofiles\Mod03 folder.

3. In Solution Explorer, expand Data Source Views and double-click Adventure Works Demo
DSV.dsv.
4. On the Data Source View menu, click Add/Remove Tables. In the Add/Remove Tables dialog box,
add the DimEmployee table to the Included objects list and click OK.

5. In the data source view diagram, right-click the DimEmployee table and click New Named
Calculation. In the Create Named Calculation dialog box, create a named calculation with the
column name Full Name and the expression [FirstName] + ' ' + [LastName], and then click OK.

6. In Solution Explorer, right-click Dimensions and click New Dimension.

7. On the Welcome to the Dimension Wizard page, click Next.

8. On the Select Creation Method page, select Use an existing table and click Next.

9. On the Specify Source Information page:

o In the Main table box, select DimEmployee.

o In the Key columns list, ensure that EmployeeKey is selected.

o In the Name column box, select Full Name.

o Click Next.
10. On the Select Related Tables page, clear the Sales Territory check box, and click Next.

11. On the Select Dimension Attributes page, click Next.

12. On the Completing the Wizard page, in the Name box, type Employee and click Finish.

13. Leave Visual Studio open for the next demonstration.

Name the Levels in a Parent-Child Hierarchy

1. In the Employee.dim Dimension Designer, on the Dimension Structure tab, in the Attributes
section, click Parent Employee Key.

2. In the Properties pane, note that the Usage property for this attribute has been set to Parent.

3. In the NamingTemplate property value, click the ellipses (…).

4. In the Level Naming Template dialog box, in the first blank Name cell, type Executive.

5. In the next blank Name cell, type Middle Management.

6. In the next blank Name cell, type Senior Employee.

 MCT USE ONLY. STUDENT USE PROHIBITED
3-12 Working with Cubes and Dimensions

7. In the next blank Name cell, type Junior Employee then click OK.

8. In the Properties pane, change the MembersWithData property value to NonLeafDataHidden.

9. In the Attributes pane, click Employee dimension, and set the UnknownMember property to None.
Select the Employee Key attribute, expand its KeyColumns property, expand the
DimEmployee.EmployeeKey column, and set the NullProcessing property to Automatic.

10. In Solution Explorer, right-click Demo and click Deploy. If prompted for a password, type Pa55w.rd
for the ADVENTUREWORKS\ServiceAcct user, and then click OK.
11. When deployment has completed, in the Employee.dim Dimension Designer, in the Browser tab,
expand the employees in the hierarchy. Note that, when you select an employee, the hierarchy level
name is displayed at the top of the browser area.

12. Leave Visual Studio open for the next demo.

Question: Excluding organizational charts, where else might you use parent-child
hierarchies?
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 3-13

Lesson 3
Implementing Sorting and Grouping for Attributes
A cube can be difficult to navigate if it has numerous attributes and attribute hierarchies. In Analysis
Services, you can sort by the member name, the member key, or by a related attribute. You can also
group attributes. A member group is a system-generated collection of consecutive dimension members.
To improve usability in Analysis Services, attribute members can be formed into member groups through
a process called discretization.

Lesson Objectives
This lesson describes sorting and grouping attributes. After completing this lesson, you will be able to:

 Sort attributes.

 Group attributes.

Sorting Attributes
You can choose any attribute by which to sort a
hierarchy. This attribute can be the key, the name,
or any other secondary attribute in the dimension
table.
 If you order by name, Analysis Services will
put the members in alphanumeric order.

 If you order by key, you can specify one or

more key columns and use these to sort the
members. This allows you to sort dates by
quarters. With a single key, the order of
quarters would be Q1, Q2, Q3, and Q4 with all
years aggregated. For example, Q1 would
aggregate all Q1 data from every year. If you add the Year column to the key, the order would be Q1
2000, Q2 2000, Q3 2000, Q4 2000, Q1 2001, Q2 2001, Q3 2001, and so on.

 You can also order by a secondary attribute, which could be a standard or calculated column created
in the data source view. For example, you might want to sort a Course dimension in which the course
name runs from “Course 1” to “Course 450”. The course key is provided by the course vendor and is
not a relevant sort order. If you sort by name, the order is “Course 1”, “Course 11”, “Course 111”,
“Course 2”, “Course 22”, and so on. Therefore, you create a new column in the dimension table with
the course number as a numeric field. This can then be used to sort data in the correct order.
 MCT USE ONLY. STUDENT USE PROHIBITED
3-14 Working with Cubes and Dimensions

Grouping Attributes
Some hierarchies have no natural levels, so
Analysis Services only creates All and Leaf levels.
This can make the cube difficult to navigate if
there are many members in the hierarchy. You can
use grouping to organize members into groups to
simplify cube browsing. For example, customers
could be grouped into income brackets rather
than showing individual incomes. You can then
drill down into the groups to show the detail as
needed.

To enable grouping, you must set the

DiscretizationMethod property of the attribute.
This can be set to:

 Equal Areas – divides members into groups of equal numbers.

 Clusters – uses a clustering algorithm to group members based on the training data; this can form
useful groups, but has a higher processing cost.

To specify the number of groups, you must set the DiscretizationBucketCount property. The default is
based on the square root of the number of distinct members.
To specify a naming template, you must set the Format option for the NameColumn property of an
attribute. The default-naming template displays the first and last group members in the format “January –
March”. You can create your own naming templates to modify group names. Because naming templates
are based on the members they contain, the names of groups can change as members are added or
removed.

Demonstration: Using Sorting and Grouping

In this demonstration, you will see how to use Sorting and Grouping.

Demonstration Steps
Use Sorting

1. In Visual Studio, in Solution Explorer, expand Dimensions, and then double-click Dim Date.dim.

2. In the Browser tab, ensure Date is selected in the Hierarchy drop-down and explore the Date
dimension, noting that the month numbers are not in numeric order.

3. On the Dimension Structure tab, in the Attributes section, click Month Number Of Year.

4. In the Properties pane, change the OrderBy property to Key.

5. On the Dimension menu, click Process. If you receive a message about out-of-date content, click
Yes, and then click Yes to confirm you want to continue with deployment. If you are prompted for a
password, type Pa55w.rd and click OK.

6. In the Process Dimension – Dim Date dialog, click Run, and when the build process has completed,
in the Process Progress dialog, click Close.

7. In the Process Dimension – Dim Date dialog, click Close.

 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 3-15

8. In the Browser tab, ensure Date is selected in the Hierarchy menu and then, on the Dimension
menu, click Reconnect.
9. Browse the Date dimension noting the month numbers are now sorted correctly.

Use Grouping

1. In Solution Explorer, expand Dimensions, and then double-click Dim Reseller.dim.

2. On the Browser tab, ensure Resellers is selected in the Hierarchy drop-down, and then browse the
Resellers hierarchy, noting the data is presented as one long list of reseller IDs. To enable simpler
location of a specific retailer ID, these will be grouped.

3. On the Dimension Structure tab, in the Attributes pane, click Reseller Key, in the Properties pane
change the DiscretizationMethod property to Automatic, and then change the
DiscretizationBucketCount to 10.

4. On the Dimension menu, click Process. If you receive a message about out-of-date content, click
Yes, and then click Yes to confirm you want to continue with deployment. If you are prompted for a
password, type Pa55w.rd and click OK.
5. In the Process Dimension – Dim Reseller dialog, click Run, and when the build process has
completed, in the Process Progress dialog, click Close.

6. In the Process Dimension – Dim Reseller dialog, click Close.

7. On the Browser tab, ensure Resellers is selected in the Hierarchy drop-down and then, on the
Dimension menu, click Reconnect.

8. Browse the Reseller dimension noting Reseller IDs are now grouped.
Question: When might you need to use a secondary attribute for sorting data?
 MCT USE ONLY. STUDENT USE PROHIBITED
3-16 Working with Cubes and Dimensions

Lesson 4
Slowly Changing Dimensions
In many dimension tables, the stored data never changes. For example, if a sale takes place in Japan, that
location is fixed. However, in other dimensions, some attributes may change. For example, the sales
representative who made that sale in Japan might move departments, offices, or work for a different
manager. With changing dimensions like this one, you should consider carefully how a cube stores
attribute values. For example, is it important to store old versions of attributes? In this lesson, you will see
how to design and implement dimensions that might change over time.

Lesson Objectives
At the end of this lesson, you will be able to:

 Describe Type 1, Type 2, and Type 3 slowly changing dimensions.

 Explain the operations that the Slowly Changing Dimension Transformation can perform on incoming
rows during a data load.

 Use the Slowly Changing Dimension Wizard to configure the Slowly Changing Dimension
Transformation.

What Are Slowly Changing Dimensions?

Many dimensions have values that do not change.
For example, if a customer purchases a product at
noon on December 12, that time does not change,
regardless of later events. However, if the sales
representative who helped the customer worked
in the Seattle office, that representative might
later move to the Paris office. Changes such as this
to the values of a dimension attribute might affect
analyses based on your cube. You should ensure
that your dimension design considers this. In this
case, you might want the sale to remain credited
to the Seattle office even after the representative
moves to Paris. This would require that both the old office and the new office are stored for the
representative, together with the date when that person moved.

Any dimension that is subject to such change is termed a Slowly Changing Dimension (SCD). You can take
three different approaches to their design:

 Type 1. These changes are the simplest type of SCD to implement. Attribute values are updated
directly in the existing dimension table row and no history is maintained. This makes Type 1 changes
suitable for attributes that are used to provide drill-through details, but unsuitable for analytical
slicers or hierarchy members where historic comparisons must reflect the attribute values as they were
at the time of the fact event.

 Type 2. These changes involve the creation of a fresh version of the dimension entity in the form of a
new row. Typically, a bit column in the dimension table is used as a flag to indicate which version of
the dimension row is the current one. Additionally, datetime columns are often used to indicate the
start and end of the period for which a version of the row was (or is) current. Maintaining start and
end dates makes it easier to assign the appropriate foreign key value to fact rows as they are loaded
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 3-17

so that they are related to the version of the dimension entity that was current at the time the fact
occurred.
 Type 3. These changes are rarely used. In a Type 3 change, the previous value (or sometimes a
complete history of previous values) is maintained in the dimension table row. This requires
modifying the dimension table schema to accommodate new values for each tracked attribute, and
can result in a complex dimension table that is difficult to manage.

The Slowly Changing Dimension Transformation

The Slowly Changing Dimension Transformation
applies the changes required by Type 1 and Type
2 SCDs during data loads. The logic required for
Type 2 SCDs in particular can become complex—
you should use this transformation to handle row
matching, detect incoming rows that require
changes, and to insert new records that are
correctly linked to old values. The Slowly Changing
Dimension Transformation also handles fixed
attributes, which should never be changed by
incoming values, and inferred members, which are
members of a dimension table that do not yet
exist but are referenced by data in a fact table.

Note: Type 3 SCDs are rarely used and not supported by the Slowly Changing Dimension
Transformation or the Slowly Changing Dimension Wizard.

The Slowly Changing Dimension Transformation has a workflow of operations that includes the following:

 A path to insert new dimension records in an SCD.

 A path to update dimension records where a changing attribute has been modified and no historical
dimension data is required. This is an implementation of a Type 1 change.

 A path to insert a new record for dimension members where a historic attribute has been modified.
This path also updates the current record indicator. This is an implementation of a Type 2 SCD.

 A path to insert minimal records for inferred members in the source data.

For more information, see Slowly Changing Dimension Transformation in Microsoft Docs:

Slowly Changing Dimension Transformation

https://aka.ms/m4r621
 MCT USE ONLY. STUDENT USE PROHIBITED
3-18 Working with Cubes and Dimensions

The Slowly Changing Dimension Wizard

Use the Slowly Changing Dimension Wizard to
configure the Slowly Changing Dimension
Transformation. This can be a complex task but
the wizard guides you through the configuration
in a straightforward manner. The wizard enables
you to configure:

 The columns containing keys that can be used

to look up existing dimension records in the
data warehouse.

 The non-key columns that are fixed, changing,

or historic attributes.

 Whether or not changes to a fixed column should produce an error or be ignored.

 The column in the dimension table that should be used to indicate the current version of a dimension
member for which historic attributes have changed over time.
 Whether or not the staged data includes inferred members for which a minimal record should be
inserted. Inferred members are identified based on the value of a specified column in the source.

Best Practice: When you design and select key columns in dimension tables, ensure that
they are unchanging. For example, in a product database, a product number or part number is
often used as a primary key because it is unique to the product. However, a product number or
part number can change and so should be an attribute of an SCD. If you want to store both old
and new product numbers for later analysis, you need a Type 2 SCD and two rows with different
product numbers. Ensure that the Slowly Changing Dimension Transformation can identify these
records as relating to the same product by using a different identity value for the primary key in
the dimension table.

Question: You are implementing a Customer dimension in a cube. Should you implement
this as a fixed dimension, a Type 1 SCD, or a Type 2 SCD?
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 3-19

Lab: Working with Cubes and Dimensions

Scenario
You have created a cube and tested it with a small set of business users whose feedback says that the
cube shows potential but is difficult to navigate intuitively. To resolve this problem, you plan to refine the
cube’s dimensions to include the attributes by which users want to aggregate data, and hierarchies that
make it easy to do so at multiple levels.

Objectives
After completing this lab, you will be able to:

 Implement dimensions in a cube.

 Configure sorting and grouping of dimensions.

Estimated Time: 90 minutes

Virtual machine: 20768C-MIA-SQL

User name: ADVENTURWORKS\Student

Password: Pa55w.rd

Exercise 1: Configuring Dimensions

Scenario
Several users at Adventure Works Cycles want to be able to analyze data across time periods. You need to
configure the Date dimension as a Time dimension so that Analysis Services can apply temporal
calculations to values.

The main tasks for this exercise are as follows:

1. Prepare the Lab Environment

2. Remove Unused Attributes

3. Add Dimension Intelligence

 Task 1: Prepare the Lab Environment

1. Ensure the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and then log
on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd.

2. Run Setup.cmd in the D:\Labfiles\Lab03\Starter folder as Administrator.

 Task 2: Remove Unused Attributes

1. Open the Adventure Works OLAP.sln solution in the D:\Labfiles\Lab03\Starter folder in Visual
Studio.

2. Open the Customer dimension in the Dimension Designer and note that many attributes have been
added to allow business users to aggregate measures in many different ways. However, users have
complained that some of these attributes are unnecessary and that they should be removed to make
browsing the cube simpler.
3. Delete the Commute Distance, Number Cars Owned, and Number Children At Home attributes
from the Customer dimension.

4. Similar feedback has been received about the Product dimension, so delete the Days To
Manufacture and Safety Stock Level attributes from the Product dimension.
 MCT USE ONLY. STUDENT USE PROHIBITED
3-20 Working with Cubes and Dimensions

 Task 3: Add Dimension Intelligence

1. Use the Business Intelligence Wizard on the Date dimension to define dimension intelligence.

2. Specify that the dimension is a Time dimension.

3. Map the dimension attribute columns as follows:

o Year: Calendar Year

o Half Year: Calendar Semester

o Quarter: Calendar Quarter

o Month: Full Date Alternate Key

o Date: Month

Results: After this exercise, unused attributes in the Customer and Product dimensions will have been
removed; time intelligence will have been added to the Date dimension.

Exercise 2: Defining Relationships and Hierarchies

Scenario
Business users need to view aggregations at different levels. Specifically, analysts want to view business
results for product categories, and then drill down to see details for subcategories and individual
products. Analysts also want to view customers by gender and then, for each gender, view business results
based on marital status.

The main tasks for this exercise are as follows:

1. Create a Natural Hierarchy

2. Create a Non-Natural Hierarchy

3. Create a Hierarchy with Attribute Relationships

 Task 1: Create a Natural Hierarchy

1. Edit the Product.dim dimension to create a hierarchy that contains the following attributes:

o English Product Category Name

o English Product Subcategory Name

o English Product Name

Note: if a warning is displayed, notifying you that attribute relationships do not exist and
performance may be decreased, ignore it. You will see how to use attribute relationships to optimize
a hierarchy later in this lab.

2. Rename the hierarchy to Categorized Products.

3. Rename the hierarchy levels as follows:

o Category

o Subcategory
o Product

4. Set the AttributeHierarchyVisible property of all attributes to False so that the Categorized
Products hierarchy is the only way to browse the Product dimension.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 3-21

5. Process the dimension when you have finished.

o If you are prompted to redeploy the project, do so.

o If you are prompted for impersonation credentials, enter the password Pa55w.rd for the
ADVENTUREWORKS\ServiceAcct account.

6. Browse the Categorized Products hierarchy in the dimension browser.

7. Save the dimension.

 Task 2: Create a Non-Natural Hierarchy

1. In the Customer dimension, create a hierarchy named Gender-Marital Status that includes the
following attributes:

o Gender

o Marital Status
Note: if a warning is displayed, saying that attribute relationships do not exist and that performance
may be decreased, ignore it. You will see how to use attribute relationships to optimize a hierarchy
later in this lab.

2. Process the dimension when you have finished:

o If you are prompted to redeploy the project, do so.

o If you are prompted for impersonation credentials, enter the password Pa55w.rd for the
ADVENTUREWORKS\ServiceAcct account.
3. Browse the Gender-Marital Status hierarchy in the dimension browser. Note that gender can be F,
M, or Unknown, and that marital status can be M or S.

4. Save the dimension when you have finished.

 Task 3: Create a Hierarchy with Attribute Relationships

Configure Attribute Column Bindings

1. In the Dimension Designer for the Date dimension, modify the Calendar Semester attribute so that a
semester is uniquely identified by both the calendar year and semester, but displayed using only the
calendar semester value. To do this:

o In the properties for the Calendar Semester attribute, set the KeyColumns property so that the
key columns are CalendarYear followed by CalendarSemester.

o Set the NameColumn property so that the source column value is CalendarSemester.

o Set the ValueColumn property so that the source column value is CalendarSemester.

2. Modify the Calendar Quarter attribute so that a quarter is uniquely identified by both the calendar
year and the calendar quarter, but displayed using only the calendar quarter value. To do this:

o In the properties of the Calendar Quarter attribute, set the KeyColumns property so that the
key columns are CalendarYear followed by CalendarQuarter.

o Set the NameColumn property so that the source column value is CalendarQuarter.

o Set the ValueColumn property so that the source column value is CalendarQuarter.
 MCT USE ONLY. STUDENT USE PROHIBITED
3-22 Working with Cubes and Dimensions

3. Modify the Month attribute so that a month is uniquely identified by both the calendar year and the
month number of year, but displayed using the month name value. To do this:
o Set the KeyColumns property so that the key columns are Calendar Year followed by
MonthNumberOfYear.

o Set the NameColumn property so that the source column value is EnglishMonthName.

o Set the ValueColumn property so that the source column value is EnglishMonthName.

Create Attribute Relationships

1. In the Date dimension, create the following attribute relationships:

o Full Date Alternate Key > Month (Rigid)

o Month > Calendar Quarter (Rigid)

o Calendar Quarter > Calendar Semester (Rigid)

o Calendar Semester > Calendar Year (Rigid)

2. Save the dimension when you have finished.

Create a Hierarchy
1. Modify the Date dimension structure to create a new hierarchy named Calendar Date that contains
the following levels:

o Calendar Year
o Calendar Semester

o Calendar Quarter

o Month
o Full Date Alternate Key (renamed to Day)

2. Set the AttributeHierarchyVisible property of the attributes that are included in the Calendar Date
hierarchy to False so that they can only be used by browsing the hierarchy.
3. Process the dimension:

o If you are prompted to redeploy the project, do so.

o If you are prompted for impersonation credentials, enter the password Pa55w.rd for the
ADVENTUREWORKS\ServiceAcct account.

Results: After this exercise, you should have created a Categorized Products hierarchy and a Gender-
Marital Status hierarchy.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 3-23

Exercise 3: Sorting and Grouping Dimension Attributes

Scenario
Having configured a Date dimension, it has been reported that months are being sorted alphabetically
rather than chronologically. In this lab, you will configure the date dimension so that months are sorted
chronologically.

The main tasks for this exercise are as follows:

1. Sort Attribute Members

2. Group Attribute Members

 Task 1: Sort Attribute Members

1. Browse the Calendar Date hierarchy in the dimension browser. Note that months are displayed in
alphabetical, rather than chronological order.

2. Modify the Month attribute to set its OrderBy property to Key. The key for this attribute consists of
the Calendar Year and Month Number of Year columns, so sorting by key should ensure that
months are displayed in chronological order.
3. Reprocess the dimension and verify that the months are now displayed in the correct order.

 Task 2: Group Attribute Members

1. Open the Customer dimension in Dimension Designer, and explore the data in the Customer table.
Notice the range of values for the YearlyIncome column.

2. Process the dimension and browse the Yearly Income hierarchy.

o If you are prompted to redeploy the project, do so.

o If you are prompted for impersonation credentials, enter the password Pa55w.rd for the
ADVENTUREWORKS\ServiceAcct account.

3. Browse the dimension and view the Yearly Income hierarchy. Note that this hierarchy has no
structure.

4. Modify the following properties of the Yearly Income attribute:

o DiscretizationMethod: Automatic

o DiscretizationBucketCount: 5

o OrderBy: Key

5. Reprocess the dimension and browse the Yearly Income hierarchy to verify that the values are now
grouped into five income bands, with a sixth band for customers whose income is unknown.

Results: You will have a multidimensional project with enhanced dimensions.

 MCT USE ONLY. STUDENT USE PROHIBITED
3-24 Working with Cubes and Dimensions

Module Review and Takeaways

Dimensions are a fundamental component of cubes. Facts or measures have the values that interest you,
but dimensions provide the business context by which you aggregate these measures. Dimensions
organize your data by category and show results grouped by these categories—such as product,
customer, or month.

Microsoft SQL Server Analysis Services dimensions contain attributes that correspond to columns in
dimension tables. These attributes appear as attribute hierarchies and can be organized into user-defined
hierarchies, or defined as parent-child hierarchies, based on columns in the underlying dimension table.
Hierarchies are used to organize measures contained in a cube.
 MCT USE ONLY. STUDENT USE PROHIBITED
4-1

Module 4
Working with Measures and Measure Groups
Contents:
Module Overview 4-1 
Lesson 1: Working with Measures 4-2 

Lesson 2: Working with Measure Groups 4-6 

Lab: Configuring Measures and Measure Groups 4-15 

Module Review and Takeaways 4-19 

Module Overview
Measures and measure groups specify the data values that your cube can aggregate to provide summary
values for analysis. This module describes measures and measure groups. It also explains how you can use
them to define fact tables and associate dimensions with measures.

Objectives
After completing this module, you will be able to:

 Configure measures.
 Configure measure groups.
 MCT USE ONLY. STUDENT USE PROHIBITED
4-2 Working with Measures and Measure Groups

Lesson 1
Working with Measures
This lesson explains how to work with measures, including configuring how measures are displayed and
aggregated, and information about measure properties.

Lesson Objectives
After completing this lesson, you will be able to:

 Describe the concept of a measure.

 List the properties of measures.

 Configure measure formatting.

 Choose appropriate aggregation functions for required additivity.

Introducing Measures
A measure represents a column that contains
quantifiable data, usually numeric, that you can
aggregate. Typically, measures occur as a column
in the fact table and are the values that the
business user finds most interesting. Measures are
organized into measure groups and are distinct
from dimensions.

For example, in a retail cube that has dimensions

for stores, products, time, and salespeople, the
measures might include sales value and units.
These measures are then aggregated to provide
total sales units and value for each store, product,
day, and salesperson.

To provide specific aggregated details, you might select a store from the stores dimension, a product from
the products dimension, a day from the time dimension, a salesperson from the salesperson dimension
and, from the measures dimension, either sales units or sales value.

You create measures as part of the Cube Wizard or on the Cube Structure tab of the Cube Designer.

Calculated Measures

You can create calculated measures based on other measures in the fact table. You use a “measure
expression” to define the value of a calculated measure, based on a fact table column as modified by a
Multidimensional Expressions (MDX) statement. You should consider the performance impact of using
calculated measures because they are worked out at run time. If a calculated measure is infrequently used
and is quite simple—for example, the sum of two measures—it will have little performance impact on
queries, but might reduce processing times. If a measure is frequently used and involves a complex
calculation, it might cause unacceptable query times.

Measures from Attribute Columns

You can use attribute columns from dimension tables to define measures. However, these are typically
semiadditive or nonadditive, which means that they cannot be fully aggregated across all dimensions. You
will learn more about semiadditive and nonadditive measures later in this lesson.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 4-3

Aggregation Function

The aggregation behavior of each measure is determined by the aggregation function that is associated
with the measure. You will learn more about aggregations later in this lesson.

Granularity

You should consider the granularity of measures, which refers to how specific something is. For example,
an hour is more granular than a day, and a day is more granular than a week. You cannot make a measure
more granular than the highest level of granularity in the dimension tables, but you can make it less so.
For example, if you have a time dimension that stores data at the levels of day, month, quarter, and year,
you cannot store sales measures at a granularity of hours, but you can store them at a granularity of
months. As you will see later in this module, you can specify the granularity of a measure group in relation
to a specific dimension.

Properties of Measures
Measures have properties that enable you to
define how the measures function and to control
how the measures appear to users. You can
configure measures on the Cube Structure tab of
the Cube Designer.

Measures inherit certain properties from the

measure group of which they are a member,
unless those properties are overridden at the
measure level.

Measure properties determine:

 How measures are aggregated.

 The data type of the column in the underlying fact table to which the measure is bound.

 The column in the data source view to which the measure is bound.

 The description of the measure, which might be exposed in client applications.

 The name that is displayed to the user.

 The folder in which the measure will appear when users connect to the cube.
 The display format of the measure.

 The unique identifier (ID) of the measure; note that this property is read-only.

 Any MDX expressions that define the measure.

 The visibility of the measure.

 MCT USE ONLY. STUDENT USE PROHIBITED
4-4 Working with Measures and Measure Groups

Configuring How Measures Are Displayed

To determine how values of a measure are
displayed to users, by using the Properties window
of the measure in the Cube Designer, you can
configure the measure property FormatString.
Typical FormatString values include Standard or
Currency, but, although a list of display formats is
provided, you can specify many additional formats
that are not in the list by specifying any named or
user-defined format that is valid in Microsoft®
Visual Basic®.

The table below shows some examples of formats

that can be specified.

Display Format Value Example Output

Standard 123456.78

Currency $123,456.78

0 123457

$#,#.00 $123,456.78

#,#0.000 123,456.789

The last two rows in the above table show user-defined formats that have been entered, rather than
chosen from the list. In this example, the regional setting in Control Panel on the client computer is
English (United States).

Note: The currency format will display the currency symbol of the client’s location.
Therefore, the value may be incorrect and, in most scenarios, it is more appropriate to employ a
user-defined FormatString property.

Aggregation Functions
Aggregation functions determine the function that
is used to aggregate the measure. Before applying
an aggregation function to a measure, you should
consider how the measure will aggregate along all
of the dimensions in the measure group that
contains the measure. This is called the additivity
of the measure, of which there are three types:

 An additive measure. An additive measure,

also called a fully additive measure, can be
aggregated along all of the dimensions
included in the measure group that contains
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 4-5

the measure, without restriction. Additive measures are the most commonly used in SQL Server
Analysis Services.
 A semiadditive measure. A semiadditive measure can be aggregated along some, but not all,
dimensions that are included in the measure group that contains the measure. An example of a
semiadditive measure is a units-in-stock value for warehouse goods. This would be additive by region
to reveal the total units in stock, but would not be additive by time because, if you add last week's
stock to this week's stock, the resulting value could be twice as much stock as you actually have.

 A nonadditive measure. A nonadditive measure cannot be aggregated along any dimension in the
measure group that contains the measure. Instead, you must individually calculate the measure for
each cube cell that represents the measure. For example, a calculated measure that returns a
percentage, such as profit margin, cannot be aggregated from the percentage values of child
members in any dimension.

By default, the AggregateFunction property is set to Sum. You can change the AggregateFunction
property for a measure in the Properties window of the measure in the Cube Designer. All aggregation
functions are additive, semiadditive, or nonadditive.

Commonly used aggregation functions include:

 The Sum function. This is additive and calculates the sum of all values for every child member. Sum is
the default function.

 The Count function. This is additive and calculates the quantity of child members.

 The Min function. This is semiadditive and calculates the lowest value for each child member.

 The Max function. This is semiadditive and calculates the highest value for each child member.

 The DistinctCount function. This is nonadditive and calculates the count of all unique child members.

 The None function. This is nonadditive, performs no aggregation, and supplies values directly from
the fact table.

Other semiadditive aggregation functions include:

 The ByAccount function. This calculates aggregation according to the aggregation function that is
assigned to the account type for a member in an account dimension.

 The AverageOfChildren function. This calculates the average of values for all non-empty child
members.
 The FirstChild function. This retrieves the value of the first child member.

 The LastChild function. This retrieves the value of the last child member.

 The FirstNonEmpty function. This retrieves the value of the first non-empty child member.

 The LastNonEmpty function. This retrieves the value of the last non-empty child member.

Question: Give an example of scenarios where you would use an additive measure, a
semiadditive measure, and a nonadditive measure.
 MCT USE ONLY. STUDENT USE PROHIBITED
4-6 Working with Measures and Measure Groups

Lesson 2
Working with Measure Groups
You use measure groups to associate dimensions with measures. Properties that you can configure at the
measure group level include:

 Partitions. These are containers for a portion of the measure group data.

 Aggregations. These are precalculated summaries of data from leaf cells that improve query
response times.

Lesson Objectives
After completing this lesson, you will be able to:

 Describe measure groups.

 List the properties of measure groups.

 Describe the relationships between measure groups and dimensions.

 Describe aggregations and their effect on performance.

 Explain partitions and measure group storage.

 Explain how to configure measure group storage.

Introducing Measure Groups

In a cube, measures are sorted into measure
groups by their underlying fact tables. You use
measure groups to associate dimensions with the
measures in the fact table.

A measure group defines:

 The dimensions by which the measures can be

aggregated. This is achieved by creating
relationships between measure groups and
dimensions in the cube. For example, you
might relate an Internet Sales measure group
to Product, Date, and Customer dimensions.
In the same cube, you might relate a Reseller
Sales measure group to Product, Date, and Reseller dimensions. Measures in both groups can be
aggregated by product and date, Internet Sales measures by customer, and Reseller Sales
measures by reseller.
 The granularity at which the measures are related to the dimensions. For example, a cube might
include a Reseller Sales measure group that is related to the Date dimension based on a key for an
individual day (because each sale is associated with the specific date on which it occurred), and a
Sales Quota measure group related to the Date dimension based on the year (because sales quotas
are set for each year, not for individual days, weeks, or months).

 How measure aggregations should be stored. For example, you might optimize cube performance by
storing preaggregated values for some or all of the possible aggregations for measures in a measure
group and its related dimensions.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 4-7

When you create a cube by using the Cube Wizard in SQL Server Data Tools, you define one or more
measure groups. By default, the Cube Wizard will create one measure group per fact table. You can add
and configure measure groups for existing cubes by using the Cube Designer.

Note: The DistinctCount aggregation looks for the number of distinct values in a fact
table column. For example, you might want to know how many sales staff sold a particular
product. This query is very resource-intensive, so SQL Server Analysis Services creates a separate
group for DistinctCount measures.

Measure Group Properties

Measure group properties determine behaviors for
the entire measure group and set default
behaviors for certain properties of measures within
a measure group. You can access the properties of
a measure group from the Properties window of
the measure group in the Cube Designer. Measure
group properties include:

 AggregationPrefix. This specifies the

common prefix that is used for all aggregation
names of the measure group.

 DataAggregation. This determines whether

SQL Server Analysis Services can aggregate
persisted data or cached data for the measure group. The default value is
DataAndCacheAggregatable, which will aggregate persisted and cached data. You can change this
to CacheAggregatable or DataAggregatable. This can be useful on systems that have limited RAM
because the other two settings will attempt to keep the aggregations in memory to improve
performance.

 ErrorConfiguration. This provides configurable error-handling settings for handling of duplicate

keys, unknown keys, null keys, error limits, action upon error detection, and the error log file.

 EstimatedRows. This specifies the estimated number of rows in the fact table. You use this when you
design aggregations. The value is static and should be kept approximately accurate, either manually
or by updating the count in the Aggregation Design Wizard.

 EstimatedSize. This specifies the estimated byte size of the measure group. You use this when you
design aggregations.

 ID. This specifies the identifier of the object.

 IgnoreUnrelatedDimensions. This determines the output when dimensions are used that are not
related to the measure group. If this is set to True, the output will show the top-level aggregation for
the dimension. If this is set to False, the output will be blank for the dimension.

 ProactiveCaching. This defines how data is reprocessed when there are changes in the underlying
data source. You will learn more about proactive caching later in this module.
 ProcessingMode. This indicates whether indexing and aggregating should occur during or after
processing. You can use lazy processing to enable SQL Server Analysis Services to choose a time when
workload is low, but you will not know when the indexing workload will occur, or when indexing has
taken place and queries are at peak performance.
 MCT USE ONLY. STUDENT USE PROHIBITED
4-8 Working with Measures and Measure Groups

 ProcessingPriority. This determines the processing priority of the cube during background
operations, such as lazy aggregations and indexing.
 StorageLocation. This is the file system storage location for the measure group. If none is specified,
the location is inherited from the cube that contains the measure group.

 StorageMode. This determines the storage mode for the measure group. You will learn more about
storage modes later in this module.

 Type. This specifies the type of the measure group.

Relationships Between Measure Groups and Dimensions

A relationship between a dimension and a
measure group consists of the dimension and fact
tables participating in the relationship and a
granularity attribute, which specifies the
granularity of the dimension in the particular
measure group. This relationship is configured on
the Dimension Usage tab of the Cube Designer.

Note: When a measure group is updated,

dimension relationship information is unavailable
until the cube is processed.

Relationship Types

There are four types of relationship between dimension and fact tables:

 Regular. A regular dimension relationship represents the relationship between dimension tables and
a fact table in a traditional star schema design. The key column for the dimension is joined directly to
the fact table.
 Reference. A reference dimension relationship represents the relationship between dimension tables
and a fact table in a snowflake schema design. The key column for the dimension is joined indirectly
to the fact table.

 Fact. If a dimension consists entirely of values in a fact table and does not have a separate dimension
table, it has a fact dimension relationship.

 Many-to-many. Typically, every dimension member relates to many facts, but all of those relate to
just one member in each dimension. This is a standard one-to-many relationship. You might also have
many-to-many relationships when, for example, a customer has multiple reasons for making a
purchase, each reason having numerous customers. In a relational database, this relationship would
be achieved through an intermediate table. In SQL Server Analysis Services, a similar technique is used
by joining the dimension to an intermediate fact table, by joining the intermediate fact table to an
intermediate dimension, and by joining the intermediate dimension to the fact table.

Note: Foreign-key relationships between all tables that are involved in a many-to-many
relationship must exist in the underlying data source view.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 4-9

Granularity

By default, a measure group will use the highest level of granularity. For example, measure groups are
often related to a date dimension, in which the highest level of granularity is day level. This is suitable for
some measure groups, but in other measure groups, the data in the fact table might not have that level of
detail, in which case you will need to change the granularity to another level, such as month or quarter. To
change the granularity of a particular relationship, you can use the Granularity attribute list.

Demonstration: Defining Relationships Between Dimensions and Measure

Groups
In this demonstration, you will see how to define a referenced relationship.

Demonstration Steps
1. Ensure that the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and then
log on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd.

2. In the D:\Demofiles\Mod04 folder, run Setup.cmd as Administrator.

3. Start SQL Server Data Tools, and then open Demo.sln in the D:\Demofiles\Mod04 folder.
4. On the Build menu, click Deploy Solution. If you are prompted, specify the user name
ADVENTUREWORKS\ServiceAcct and the password Pa55w.rd.

5. Wait for the Deployment Completed Successfully message in the Deployment Progress – Demo
window, and then close the Deployment Progress – Demo window.
6. In Solution Explorer, double-click SalesDemo.cube, and on the Browser tab, expand Measures,
expand Reseller Sales, and then drag the Revenue measure to the query results area. Expand the
Geography dimension, and then drag Country-Region to the left of the Revenue value. Note that
the values for each region are the same. The aggregation is incorrect.

7. In Solution Explorer, expand dimensions, right-click Geography.dim, and then click View Designer.
Note that this dimension is based on the Geography table and has a Country-Region attribute.

8. In Solution Explorer, right-click Reseller.dim, and then click View Designer. Note that this dimension
is based on the DimReseller table, which includes a GeographyKey attribute that relates it to the
Geography table.

9. Click the tab for the SalesDemo cube, and on the Cube Structure tab, in the Data Source View pane,
note that there is no direct relationship between the Reseller Sales fact table and the Geography
dimension table.
10. On the Dimension Usage tab, click the intersection of the Reseller Sales measure group and the
Geography dimension, and then click the ellipsis (…) button.

11. In the Select relationship type list, select Referenced.

12. In the Intermediate dimension list, select Reseller.

13. In the Reference dimension attribute list, select Geography Key, in the Intermediate dimension
attribute list, select Geography Key, and then click OK.

14. On the Dimension Usage tab, in the Dimensions list, right-click Geography, and then click
Rename.
15. Change the name of this cube dimension to Reseller Geography, and then press Enter to make this
name change take effect.
 MCT USE ONLY. STUDENT USE PROHIBITED
4-10 Working with Measures and Measure Groups

16. On the Build menu, click Deploy Solution. If you are prompted, specify the user name
ADVENTUREWORKS\ServiceAcct and the password Pa55w.rd.
17. Wait for the Deployment Completed Successfully message in the Deployment Progress – Demo
window, and then close the Deployment Progress – Demo window.

18. On the Browser tab for the SalesDemo cube, click the Reconnect button.

19. Right-click the query results area, and then click Clear Grid.

20. Expand Measures, expand Reseller Sales, and then drag the Revenue measure to the query results
area. Expand the Reseller Geography dimension, and then drag Country-Region to the left of the
Revenue value. Note that the values for each region are now correct.

Aggregations
Aggregations are precalculated summaries of leaf
cell data that improve query response time by
preparing the answers before the questions are
asked.

A simple Aggregation object consists of:

 Basic information that includes the name of
the aggregation, the ID, annotations, and a
description.
 A collection of AggregationDimension
objects that contain the list of granularity
attributes for the dimension.

Aggregations give multidimensional cubes their performance benefits. They precalculate the answer to
queries at processing time so that, when the query is run, performance is hugely improved. If you consider
that a query might have to aggregate thousands of individual values to produce one summarized output
value, of which there may be hundreds, it becomes clear that precalculating these results will have
enormous benefits.

It might seem appropriate to precalculate every possible result, but some aggregations provide little
benefit. SQL Server Analysis Services incorporates a sophisticated algorithm to select aggregations for
precalculation so that other aggregations can be quickly computed from the precalculated values. For
example, if SQL Server Analysis Services stores aggregates at a month level, it takes little processing to
generate quarter data at query run time. The Aggregation Design Wizard gives options for you to specify
storage and percentage constraints on the algorithm to achieve a satisfactory trade-off between query
response time and storage requirements. Typically, there is little benefit from preprocessing more than 30
percent of aggregations.
After using the Aggregation Design Wizard, you can further improve aggregation design for your specific
system by using SQL Server Analysis Services with normal workloads for some time, and then running the
Usage-Based Optimization Wizard. This wizard is almost identical to the Aggregation Design Wizard, but
bases optimization on queries that are submitted to SQL Server Analysis Services and should, therefore,
give better query response with less processing time.

To use the Aggregation Design Wizard or the Usage-Based Optimization Wizard for a measure group, in
the Cube Designer, on the Aggregations tab, right-click the measure, and then click Design
Aggregations or Usage Based Optimization.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 4-11

Partitions
A partition is a container for all, or a portion, of
the measure group data. When a measure group is
created, it will have one partition. You can then
add further partitions to the measure group.

SQL Server Analysis Services Enterprise Edition

uses partitions to manage and store data and
aggregations for a measure group in a cube. By
splitting the data into multiple partitions, there are
performance benefits:
 You can speed up cube updates by using an
incremental update. An incremental update is
where one partition is updated rather than
the whole measure group. For example, if you had a different partition for each month, and are only
updating the most recent month, you only need to process the latest partition, which reduces
processing time.
 You can store frequently used data in a different way from aggregated data. This means you can use
higher levels of aggregation and a faster storage mode to improve query times for the frequently
used data, while reducing processing times and storage requirements for older data.

 Partitions enable the source data and aggregate data of a cube to be distributed across multiple hard
drives and among multiple server computers. You can use this to greatly improve performance on
large cubes.

Note: When you incrementally update a partition, a temporary partition is created that has
an identical structure to the source partition. The temporary partition is then merged with the
original partition. There are possible scenarios that could affect data integrity when you
incrementally update individual partitions. For example, if you had a different partition for each
month, and the date of a record in the source data changed to or from the most recent month,
the data may either disappear or be duplicated. You should consider these scenarios and plan
accordingly.

A simple Partition object consists of:

 Basic information that includes the name of the partition, in addition to the storage and processing
modes.

 A slicing definition that is an MDX expression that specifies a tuple or a set.

 An aggregation design that is a collection of aggregation definitions that you can share across
multiple partitions.

Typically, you can partition a measure group horizontally or vertically.

In a horizontally partitioned measure group, each partition is based on a separate fact table that can be
from different data sources.

A vertically partitioned measure group is based on a single table, with each partition centered on a source
system query that filters data for the partition.
 MCT USE ONLY. STUDENT USE PROHIBITED
4-12 Working with Measures and Measure Groups

Configuring Measure Group Storage

There are three storage modes in SQL Server
Analysis Services: relational online analytical
processing (ROLAP), multidimensional online
analytical processing (MOLAP), and hybrid online
analytical processing (HOLAP).

ROLAP
ROLAP storage stores the aggregations in indexed
relational data, along with the source data. It does
not use multidimensional storage, so ROLAP data
is slower to query and process than MOLAP or
HOLAP, but it enables real-time access to data and
uses less storage space.

MOLAP

MOLAP is the most commonly used storage mode. It stores both the aggregations and a copy of the
source data in the multidimensional cube. MOLAP gives the greatest performance, but requires more
storage space due to the duplication. There is latency when you use MOLAP storage because the cube
data is refreshed only when the cube is processed, so changes from the data source are only updated
periodically.
You can use proactive caching with MOLAP storage. When proactive caching is used, the server is notified
when changes are made to the data source, and then the changes are incorporated. While the cache is
rebuilt with new data, you can choose to send queries to ROLAP data, which is up to date, but slower, or
to the original MOLAP storage, which is faster, but will not have the new data.

If you only need daily updates, you can set processing to occur every 24 hours. You can schedule
processing for when the cube is rarely used, for example, at night—however, some systems are in use 24
hours a day, seven days a week.

For maximum performance, you can use MOLAP without proactive caching or scheduling. This is best
suited to cubes that do not need the most up-to-date data. When objects are updated, some downtime
will be required. You can minimize this by processing the cube to a staging server, and using database
synchronization to copy the processed data to the production server.

HOLAP

HOLAP storage stores aggregations in the multidimensional cube and leaves source data in the relational
database. This can provide a good compromise when leaf-level data is rarely accessed. If leaf-level data is
frequently accessed, MOLAP would provide improved performance.

Note: You should also consider tabular data model storage, particularly as an alternative to
ROLAP and HOLAP. The tabular data model is covered in more detail in Module 7 of this course,
“Implementing a Tabular Data Model by Using Analysis Services.”
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 4-13

Storage Settings

You can configure storage at partition level by using the Storage Settings dialog box. This enables you to
maximize efficiency by using configurations that are most suitable for each partition.

The Storage Settings dialog box contains the following standard storage configurations, which are
combinations of storage mode and caching options:

 Real Time ROLAP. Use to set to ROLAP storage mode.

 Real Time HOLAP. Use to set to HOLAP storage mode.

 Low Latency MOLAP. Use to set to MOLAP storage mode with proactive caching and a latency
target of 30 minutes after first change—queries revert to ROLAP during update.

 Medium Latency MOLAP. Use to set to MOLAP storage mode with proactive caching and a latency
target of four hours after first change—queries revert to ROLAP during update.

 Automatic MOLAP. Use to set to MOLAP storage mode with proactive caching and a latency target
of two hours after first change—during update, queries use MOLAP data that was cached before
update started.
 Scheduled MOLAP. Use to set to MOLAP storage mode with updates scheduled every 24 hours.

 MOLAP. Use to set MOLAP without proactive caching or scheduled updates.

You can use the Storage Settings dialog box to apply these settings to existing partitions, default new
partitions in a measure, and default new measure groups within a cube.

 To configure storage for an existing partition, in the Cube Designer, on the Partitions tab, expand
the measure group, right-click the partition, and then click Storage Settings. This displays the
Storage Settings dialog box for the selected partition.

 To configure default storage settings for new partitions that are added to a measure group, in the
Cube Designer, on the Partitions tab, expand the measure group, and then click the Storage
Settings link for that measure group. This displays the Storage Settings dialog box for the selected
measure group.

 To configure default storage settings for new measure groups that are added to a cube, in the Cube
Designer, on the Cube Structure tab, in either the Measures or Dimensions pane, click the cube
object. In the Properties window, click the ellipsis (…) button for the ProactiveCaching property
setting. This displays the Storage Settings dialog box for the selected cube.
 MCT USE ONLY. STUDENT USE PROHIBITED
4-14 Working with Measures and Measure Groups

Check Your Knowledge

Question

You are configuring the storage settings

for a measure group. The data in the
group does not need to be updated
regularly. However, the data must be
available at all times, and queries that
access the data must execute with the
best possible performance. What is the
most suitable storage setting to apply to
the measure group?

Select the correct answer.

Real Time ROLAP

Real Time HOLAP

MOLAP

Automatic MOLAP

Medium Latency MOLAP

 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 4-15

Lab: Configuring Measures and Measure Groups

Scenario
Business users are analyzing Internet sales by using the multidimensional cube that you have created, but
they also need to analyze reseller sales. You have reseller sales data in the data warehouse on which the
cube is based, so you plan to add a second measure group that contains reseller measures.

Users have also requested that you remove some measures that they do not need to use when analyzing
data, and also ensure that the remaining measures are clearly named.

Users specifically need to analyze reseller sales by product, so you must ensure that the cube supports the
required relationship between the new reseller sales measure group and the Product dimension.

Finally, the data center administrator is concerned about the disk space that the cube uses, but your users
are worried about performance when they analyze Internet sales. You must optimize the storage and
aggregations of the cube to balance these concerns.

Objectives
After completing this lab, you will be able to:

 Configure measures.

 Define a regular relationship.

 Configure measure group storage.

Estimated Time: 60 minutes

Virtual machine: 20768C-MIA-SQL

User name: ADVENTUREWORKS\Student

Password: Pa55w.rd

Exercise 1: Configuring Measures

Scenario
You are refining the online analytical processing (OLAP) cube for your company. Users can currently view
measures for Internet sales, but you have additional reseller sales data in your data warehouse that they
would like to analyze. When you add the reseller sales measures, your users want you to remove any
measures that are not required for business analysis, and ensure that measures in the cube are clearly
named.

The main tasks for this exercise are as follows:

1. Prepare the Lab Environment

2. Create a Measure Group

3. Modify Measure Groups

 Task 1: Prepare the Lab Environment

1. Ensure that the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and then
log on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd.

2. Run Setup.cmd in the D:\Labfiles\Lab04\Starter folder as Administrator.

 MCT USE ONLY. STUDENT USE PROHIBITED
4-16 Working with Measures and Measure Groups

 Task 2: Create a Measure Group

1. Start SQL Server Data Tools and open the Adventure Works OLAP.sln solution in the
D:\Labfiles\Lab04\Starter folder.

2. In the Adventure Works DSV data source view, add the FactResellerSales table, together with all
related tables.

3. In the Sales cube, add a new measure group named Reseller Sales based on the FactResellerSales
table.

4. Review the names of the measures in the Reseller Sales measure group. Note that, when the
Reseller Sales measure group was set up, measures were created for all of the numerical fields in the
FactResellerSales table.

 Task 3: Modify Measure Groups

1. Delete the following measures from the cube:

o Revision Number
o Unit Price

o Extended Amount

o Unit Price Discount Pct

o Discount Amount

o Product Standard Cost

o Tax Amt
o Freight

Tip: click the Show Measures Grid icon to view all of the measures in the cube as a grid. In this view,
you can select several measures at the same time by holding down the Ctrl key.
2. Rename the following measures:

o Order Quantity (rename to Reseller Order Quantity).

o Total Product Cost (rename to Reseller Cost).

o Sales Amount (rename to Reseller Revenue).

o Fact Reseller Sales Count (rename to Reseller Sales Count).

Results: After this exercise, you should have created a new measure group for the FactResellerSales
table, removed unrequired measures, and renamed measures.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 4-17

Exercise 2: Defining a Regular Relationship

Scenario
Your users have specifically stated that they must be able to analyze reseller sales by product, so you must
create the required relationships to support aggregating reseller sales measures by the Product
dimension.

The main tasks for this exercise are as follows:

1. View Existing Dimensions for Measure Groups

2. Create a Dimension

 Task 1: View Existing Dimensions for Measure Groups

1. Deploy the Adventure Works OLAP solution, typing the user name
ADVENTUREWORKS\ServiceAcct and the password Pa55w.rd if you are prompted.

2. Browse the Sales cube and review the measures and dimensions that are available by selecting each
measure group:

o When the Internet Sales measure group is selected, the Customer dimension should be listed.

o When the Reseller Sales measure group is selected, there is no Customer dimension, because
reseller sales are sold to resellers, which are defined in a different dimension table.

 Task 2: Create a Dimension

1. Add a dimension named Reseller based on the DimReseller table to the solution:
o Do not include any related tables.

o Include only the Reseller Key, Business Type, and Reseller Name attributes.

2. On the Dimension Usage tab of the Cube Designer for the Sales cube, verify that a regular
relationship has been defined between the Reseller Sales measure group and the Reseller
dimension.

3. Deploy the Adventure Works OLAP solution, and then browse the cube, verifying that you can view
reseller revenue by attributes of the Reseller dimension.

Results: After this exercise, you should have added a Reseller dimension that uses a regular relationship
with the Reseller Sales measure group to enable you to analyze reseller sales data.
 MCT USE ONLY. STUDENT USE PROHIBITED
4-18 Working with Measures and Measure Groups

Exercise 3: Configuring Measure Group Storage

Scenario
Users have requested that the cube is optimized to improve performance when analyzing Internet sales.
However, you must balance this need for optimization against the space that is required to store
aggregations.

The main tasks for this exercise are as follows:

1. Configure Proactive Caching

2. Design Aggregations

 Task 1: Configure Proactive Caching

1. Configure proactive caching for the Internet Sales measure group in the Sales cube. Specify that the
Automatic MOLAP setting should be used.

2. Configure proactive caching for the Reseller Sales measure group in the Sales cube. Specify that the
Automatic MOLAP setting should be used.

 Task 2: Design Aggregations

1. Use the Aggregation Design Wizard to design aggregations for the Internet Sales measure group in
the Sales cube:

o Set all aggregation usage to default before starting the configuration.

o Use the wizard to count the objects in the measure group.

o Generate aggregations until the performance gain reaches 30 percent.

o Name the aggregation that you have generated as InternetSalesAgg.

2. Use the Aggregation Design Wizard to design aggregations for the Reseller Sales measure group,
using the same settings as the Internet Sales measure group.

3. Deploy the Adventure Works OLAP solution.

4. Use SQL Server Management Studio to connect to the localhost instance of SQL Server Analysis
Services, and then verify that the aggregation designs were deployed by using the Sales cube in the
Adventure Works OLAP database.

Results: After this exercise, you should have configured proactive caching and defined the storage mode
aggregations for the Internet Sales and Reseller Sales measure groups.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 4-19

Module Review and Takeaways

In this module, you have learned how to create and configure measure groups in a cube, and define
relationships between measure groups and dimensions.
MCT USE ONLY. STUDENT USE PROHIBITED
 
 MCT USE ONLY. STUDENT USE PROHIBITED
5-1

Module 5
Introduction to MDX
Contents:
Module Overview 5-1 
Lesson 1: MDX Fundamentals 5-2 

Lesson 2: Adding Calculations to a Cube 5-10 

Lesson 3: Using MDX to Query a Cube 5-17 

Lab: Using MDX 5-24 

Module Review and Takeaways 5-28 

Module Overview
When you have a relational database that contains structured data in rows and columns, you can use
Transact-SQL to write simple and effective queries to return the data that interests you. However,
Transact-SQL is not designed to handle data that has more than two dimensions, such as that found in an
online analytical processing (OLAP) cube. Instead, you can use the Multidimensional Expressions (MDX)
language both to query multidimensional data and to return data in multiple dimensions. In this module,
you will see how to use MDX to add features to a cube and display data in client applications such as SQL
Server® Reporting Services or Microsoft® Excel®.

Objectives
After completing this module, you will be able to:

 Write a simple MDX query that establishes cube context and specifies both query and slicer axes.

 Use MDX to add calculated members, named sets, and scoped assignments to cubes.

 Use MDX from SQL Server Management Studio, Excel, SQL Server Reporting Services, and custom
code to access and display data.
 MCT USE ONLY. STUDENT USE PROHIBITED
5-2 Introduction to MDX

Lesson 1
MDX Fundamentals
A table in Microsoft SQL Server consists of two-dimensional data that is organized into columns and rows.
An OLAP cube can have many more dimensions and multiple members of each dimension. The Transact-
SQL language that is used to query tabular data is not ideal for more than two dimensions, so SQL Server
Analysis Services includes the MDX query language. By using MDX, you can write queries that are easy to
read and understand, but execute against multidimensional cube data. In this lesson, you will learn the
essential features of the MDX language.

Lesson Objectives
After completing this lesson, you will be able to:

 Describe how MDX differs from Transact-SQL and how it enables developers to query
multidimensional data more easily.

 Describe the structure of a simple MDX SELECT query.

 Use SELECT and WHERE to specify query and slicer axes in an MDX query.

 Use FROM, CREATE SUBCUBE, LOOKUPCUBE, and FILTER to establish the cube context for a query.
 Use SQL Server Management Studio to issue simple MDX queries.

What Is MDX?
Most databases store data in two-dimensional
tables that consist of rows and columns. Transact-
SQL is designed to formulate queries against this
two-dimensional data and is highly efficient and
readable. However, in an OLAP cube, data that has
three or more dimensions is frequently stored.
Very complex Transact-SQL would be required to
query this data and it would be difficult for
developers to read and understand such queries.
Instead of using Transact-SQL, you can use MDX
to query OLAP cubes. In MDX, it is simple to drill
down into multiple dimensions and hierarchies,
and select pertinent data from many dimensions.

In some ways, MDX is similar to Transact-SQL. For example, SELECT statements and WHERE clauses are
common. However, these familiar elements have a different meaning and accept a different syntax in
MDX.

An MDX query usually selects a subset of all of the data in the cube and this subset frequently has a
smaller number of dimensions. For example, it is common to use MDX to return two-dimensional table-
like data that can be displayed in an Excel spreadsheet. However, you can return any number of
dimensions in your result set, up to the number of dimensions in the cube itself.

Note: There is also a hard limit of 128 query axes in an MDX SELECT statement.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 5-3

An MDX query that returns a single value is called an MDX expression. A query that returns a large set of
data, such as a table or a cube, is called an MDX statement. The following terms are also frequently used
in MDX documentation:

 Cell. A cell is the space at the intersection between the specified member of the measures dimension
and a specified member of one or more other dimensions. You can think of a cell as containing a
single piece of data.

 Tuple. A tuple uniquely identifies a slice of data from a cube. It is defined as a combination of
dimension members from different hierarchies. For example, a tuple might be all of the customers
that are located in North America.

 Set. A set is a collection of tuples that defines multiple cells from the same dimensions. For example, a
set might be all of the customers that are located in North America and all of the customers that are
located in Africa.

MDX is used in many scenarios. For example, you can use it to select data for display in Excel and SQL
Server Reporting Services. You can also use it in the definition of a cube, for example, to define a
calculated member.

Basic MDX Query Syntax

The most common form of MDX query is a SELECT
statement. This is used to return a subset of the
data in the cube for display or for use in
calculations. A SELECT statement frequently
includes the following elements:

 A SELECT clause. The SELECT clause specifies

what will appear along each axis of the result
set. For example, it might say that specific
years will appear on the columns and sales
amounts will appear on the rows. The values
that appear in each cell will be the
intersections of those years and amounts.
 A FROM clause. The FROM clause specifies the cube to execute the query against.

 A WHERE clause. The WHERE clause is optional. You can use it to restrict the data that is returned.
For example, it might specify that only data for the South West region is returned.

Navigating Hierarchies
In a cube, the members of dimensions and measures are hierarchical. For example, in a time dimension,
you might want to see data for 2016, for May 2016, or for May 1, 2016. There might also be parallel
hierarchies. For example, you may want to see data for the 2015–2016 fiscal year instead of the calendar
year. In MDX, you navigate these hierarchies by using dots. For example, to specify May 2016, you might
use the term “[2016].[Q1].[May]”. The levels in the hierarchy depend on the structure of the cube that you
are executing queries against.

Using Square Brackets

You need not always place square brackets around the name of an object in MDX. However, there are
three cases where square brackets are required:

 When the object name contains a space or another special character.

 When the object name matches an MDX keyword, such as SELECT or FROM.
 MCT USE ONLY. STUDENT USE PROHIBITED
5-4 Introduction to MDX

 When the object name begins with a numeric character, such as the name of a year.

Many MDX developers always use square brackets as a convention to avoid errors and ambiguity.

Avoiding Ambiguity
In a cube, many members might share the same name. For example, in a time dimension, Q1 might refer
to the first quarter of 2015 or the first quarter of 2016, or any other year in a single dimension. Similarly,
the name Tailspin Toys may appear in both a Suppliers dimension and a Customers dimension. It is
important to ensure that you specify objects or ranges of objects without any ambiguity to ensure that
the data displayed is what you intended.
To eliminate these ambiguities, ensure that you qualify each object name to uniquely identify it. For
example, to specify the first quarter of 2015, use [2015].[Q1]. To specify Tailspin Toys in the Customers
dimension, use [Customers].[Tailspin Toys], and so on. The precise qualification that you use will
depend on the structure and hierarchy levels in your cube.

If an object is identified ambiguously, MDX does not return an error, but returns the first occurrence of
the object in the database. For example, if there were members named Q1 for every year from 2011,
[2011].[Q1] would be returned because it is the first occurrence. Even if this is what you intended, it is
best to uniquely identify a member for the benefit of other query writers.

Query and Slicer Axes

In an MDX query, query axes specify the members
that appear in the edges of the result set. The
slicer axis filters the result set. By using a
combination of these axes, it is simple to drill
down into very specific summaries of the data in
your cube.

Using Query Axes

If you want a query that returns two-dimensional,
tabular data, you must define what appears along
the top, and down the side, of that table. These
are the query axes. If your query is intended to
return data in more than two dimensions, you
must specify more than two query axes—one for every dimension of the result set.

Note: It is very common to specify two query axes because the tabular data that is
produced can be easily read, for example, in a tabular report.

In the SELECT query, use the ON keyword to specify the axis on which to display a set of members.

In the following example, the query axes are specified by using the aliases COLUMNS, ROWS, and PAGES:

Using Axis Aliases

SELECT
{[Date].[Calendar Year].[2015], [Date].[Calendar Year].[2016]} ON COLUMNS,
{[Measures].[Sales Amount]} ON ROWS,
{[Customers].[Tailspin Toys], [Customers].[Wingtip Toys]} ON PAGES
FROM [Adventure Works Cube]
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 5-5

In a SELECT query, you can specify up to 128 query axes. The first five of these axes have the aliases
COLUMNS, ROWS, PAGES, SECTIONS, and CHAPTERS. Instead of using these aliases, you can use an
ordinal number, starting from 0 for the columns.

In the following example, the query axes are specified by using the ordinal numbers:

Using Axis Ordinal Numbers

SELECT
{[Date].[Calendar Year].[2015], [Date].[Calendar Year].[2016]} ON 0,
{[Measures].[Sales Amount]} ON 1,
{[Customers].[Tailspin Toys], [Customers].[Wingtip Toys]} ON 2
FROM [Adventure Works Cube]

Note: You cannot skip axes in a SELECT query. For example, you cannot specify axes 0 and
2 without specifying axis 1. Similarly, you cannot use a PAGES axis without using both a
COLUMNS axis and a ROWS axis.

In the above example, the 0 axis will include a 2015 column and a 2016 column. To return a column for
every calendar year in the Date dimension, you could use the Members function. For example, use
[Date].[Calendar Year].Members.

Using a Slicer Axis

A slicer axis filters the data that the SELECT expression returns. It is defined by the WHERE clause in an
MDX SELECT query and it specifies a condition that must be met by all of the data that is summarized in
the cells of the result set.
The following example is the same as the previous queries except that sales amounts will only be summed
if they were made by the sales representative, Nicholas Rose:

Using a Slicer Axis

SELECT
{[Date].[Calendar Year].[2015], [Date].[Calendar Year].[2016]} ON COLUMNS,
{[Measures].[Sales Amount]} ON ROWS,
{[Customers].[Tailspin Toys], [Customers].[Wingtip Toys]} ON PAGES
FROM [Adventure Works Cube]
WHERE [Sales Reps].[Name].[Nicholas Rose]

A slicer axis does not filter the members that appear along the edges of the result set. For example, it does
not affect the rows that are returned. Instead, it filters the cube cells that will be summed (or otherwise
aggregated) to generate the result cells.
 MCT USE ONLY. STUDENT USE PROHIBITED
5-6 Introduction to MDX

Establishing Cube Context

Before you execute a query, you must define what
the data query will run against. This definition is
known as the cube context. Often, you establish
the cube context simply by specifying the OLAP
cube, but, in this topic, you will also see methods
that you can use to narrow or broaden that
context.

Using the FROM Clause

In MDX, you specify the cube to execute a query
against by using the FROM clause.

In the following example, the cube context is the

entire Adventure Works cube:

Using the FROM Clause

SELECT
{[Date].[Calendar Year].[2015], [Date].[Calendar Year].[2016]} ON COLUMNS,
{[Measures].[Sales Amount]} ON ROWS
FROM [Adventure Works]

Creating a Subcube
Sometimes, you might be interested in executing multiple MDX queries against a small portion of the
cube. For example, you might want to investigate several aspects of sales performance in North America
during 2015. In such cases, you can, of course, specify query and slicer axes individually for each query,
but it is more efficient to create a subcube of the data that interests you, and then execute queries against
it.

Use the CREATE SUBCUBE statement to restrict a cube in this way. You pass to this statement an MDX
expression that defines the content of the subcube.

In this example, a subcube is created that includes sales amounts for all calendar years and all customers.
Other dimensions in the original cube are left out of the subcube.

Creating a Subcube
CREATE SUBCUBE [Adventure Works] AS
SELECT
{[Date].[Calendar Year].Members} ON COLUMNS,
{[Measures].[Sales Amount]} ON ROWS,
{[Customers].Members} ON PAGES
FROM [Adventure Works]

Note: In the previous example, the subcube shares the same name as the original cube
(“Adventure Works”). All subsequent MDX queries that have “Adventure Works” in the FROM
clause will execute against the subcube until the session closes, or until you issue a DROP
SUBCUBE statement.

Note that some members of a cube might be present in the subcube even though you did not explicitly
specify them in the CREATE SUBCUBE statement. For example, the subcube will include all of the
descendants and ascendants of the members that you specified.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 5-7

Also, be careful to understand aggregation in subcubes. For example, consider a subcube that includes
the US states, Washington and Oregon, plus the country, the United States. The reported sales for the
United States member will be the sum of sales in Washington and Oregon, but sales from other states will
not be added, because those states are not included in the subcube.

Using the LOOKUPCUBE Function

In addition to narrowing the cube context by using subcubes, you can broaden the cube context by using
the LOOKUPCUBE function. This function enables you to include members from cubes other than the one
that you specify in the FROM clause. All cubes in your query must be in the same database.

In the following example, the “Sales Amount” measure from the “Adworks Archive” cube is added as a
new measure member named “Archive Sales Amount.” This is displayed on its own row.

Using the LOOKUPCUBE Function

WITH MEMBER [Measures].[Archive Sales Amount] AS
LOOKUPCUBE("Adworks Archive", "[Measures].[Sales Amount]")
SELECT
{ [Date].[Calendar Year].Members } ON COLUMNS,
{ [Measures].[Sales Amount], [Measures].[Archive Sales Amount] } ON ROWS,
FROM [Adventure Works]

Best Practice: Although they are useful, queries that use the LOOKUPCUBE function to
integrate two cubes are likely to perform less well than queries against a single cube. If you find
that you use LOOKUPCUBE often, consider redesigning your cubes to include all of the relevant
data.

Using the FILTER Function

The FILTER function returns a set of members where a specific condition is true. For example, if you had a
dimension member that consisted of all of the years since 1980, you could use a FILTER function to return
only the years after 2010. You can use FILTER functions in several places in an MDX query. To help to
establish cube context, you can use FILTER functions to select members to place on query axes.

In the following example, the FILTER function removes all calendar years before 2010 from the result set:

Use FILTER in Query Axes

SELECT
{ FILTER( [Date].[Calendar Year].Members, [Date].[Calendar Year]>2009 ) } ON COLUMNS,
{ [Measures].[Sales Amount] } ON ROWS,
FROM [Adventure Works]

Demonstration: Querying a Cube

In this demonstration, you will see how to write and execute MDX queries in SQL Server Management
Studio.

Demonstration Steps
Deploy and Query a Cube
1. Ensure that the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and then
log on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd. In the
D:\Demofiles\Mod05\Starter folder, run Setup.cmd as Administrator.
2. On the Start screen, type SQL Server Data Tools 2015, and then press Enter.
 MCT USE ONLY. STUDENT USE PROHIBITED
5-8 Introduction to MDX

3. In SQL Server Data Tools, on the File menu, point to Open, and then click Project/Solution.

4. In the Open Project dialog box, browse to the D:\Demofiles\Mod05\Starter folder, and then
double-click Adventure Works OLAP.sln.

5. On the Build menu, click Deploy Solution.

6. If the Microsoft Visual Studio dialog box appears, click Yes.

7. If you are prompted, specify the user name ADVENTUREWORKS\ServiceAcct and the password
Pa55w.rd, and then click OK.
8. Wait for the Deployment Completed Successfully message in the Deployment Progress –
Adventure Works OLAP window, and then close the Deployment Progress – Adventure Works OLAP
window.

9. Close Microsoft Visual Studio®.

10. Start SQL Server Management Studio.

11. In the Connect to Server dialog box, in the Server type drop-down list, select Analysis Services.

12. In the Server name text box, select MIA-SQL, and then click Connect.
13. In Object Explorer, expand Databases, right-click Adventure Works OLAP, point to New Query,
and then click MDX.

14. In the query pane, type the following code:

SELECT
{
[Measures].[Internet Sales Count],
[Measures].[Reseller Sales Count]
} ON COLUMNS,
{
[Order Date].[Calendar Date].[Calendar Year]
} ON ROWS
FROM [Sales]

15. Click Execute, and then examine the results.

16. To specify a slicer axis, add the following line of code to the previous query:

WHERE [Customer].[State Province Name].&[Ohio]

17. Click Execute, and then examine the results.

18. Close SQL Server Management Studio, without saving any changes.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 5-9

Check Your Knowledge

Question

You want to run 10 queries against the

Adventure Works cube, but you are only
interested in summarizing results from
2015 and 2016. Which of the following
techniques should you use to execute
queries most efficiently?

Select the correct answer.

Specify the 2015 and 2016 members

in each SELECT query.

Use the FILTER function in each query

to display only the 2015 and 2016
members.

Use the LOOKUPCUBE function.

Create a subcube that includes only

the 2015 and 2016 members.

Specify a slicer axis by using the

WHERE clause.
 MCT USE ONLY. STUDENT USE PROHIBITED
5-10 Introduction to MDX

Lesson 2
Adding Calculations to a Cube
In previous modules, you have seen how to create dimensions and measures and use them to define the
data in a cube. However, not all of the members of a cube must come from the underlying data source,
such as a data warehouse. In this lesson, you will see how to create additional features in a cube, including
calculated members, named sets, and other script commands. All of these features are added to the cube
by using the Calculations tab of the Cube Designer. You will also see some MDX functions that can be
very helpful when defining script commands.

Lesson Objectives
After completing this lesson, you will be able to:

 Describe calculated members and explain how they can be used to enhance the information in a
cube.

 Define named sets and explain how they can make it easier to write MDX queries.

 Describe how scoped assignments limit the cells to which a script action applies.

 Use MDX functions to locate the family members for an object, such as a parent or the first child.
 List other commonly used MDX functions and choose a function for a given calculation.

 Add calculated members to a cube by using SQL Server Data Tools.

Calculated Members
A calculated member is a cube member whose
value is calculated when a user queries the cube
(in other words, at run time) by using an MDX
expression that you define as part of the cube.
Desktop applications and MDX clients can access
the calculated member just as they access other
members of dimensions or measures. A calculated
member can be defined as a combination of cube
data, arithmetic operations, numbers, and
functions.

For example, if your cube includes the measures

for [Sale Price] and [Production Cost], you could
define a member named [Profit] as follows:

An Example of a Calculated Member Expression

( [Measures].[Sale Price] – [Measures].[Production Cost] ) / [Measures].[Sale Price]

Creating a Calculated Member

Use the Cube Designer to add a calculated member to a cube. On the Calculations tab, in the Calculation
Expressions pane, the default CALCULATE command specifies that the measures in the cube should be
aggregated according to their defined aggregate function. To define a new calculated member, leave the
default script in place and click the New Calculated Member button. In addition to the MDX expression
for the new calculated member, you must fix the following properties:
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 5-11

 Name. Specify a descriptive and unique name to identify the new member. Ensure that the name is
meaningful to users who will be browsing the cube.

 Parent hierarchy. Specify the dimension that the member appears within. If you place the calculated
member in the Measures hierarchy, it is known as a calculated measure.

 Format string. You can select how the value is displayed. For example, you could select Currency or
Percent.

 Visibility. If you want users to be able to see the calculated member as they browse the cube, select
True.

 Non-empty behavior. This is an important property for performance. See the next section for details.

Calculated Member Performance

The values of calculated members are not stored with the cube, so they do not increase the disk space
that a cube consumes. However, because calculated members are executed at run time, they can increase
demands on the processing resources of your database server. If you use many complex calculated
members, they might have a negative impact on query performance.

One way to help to reduce the processing demands of a calculated member is to carefully specify a Non-
empty behavior value. If this value is blank, SQL Server Analysis Services must evaluate the full expression
to determine whether the calculated member is empty. If, by contrast, you specify one or more members
in the Non-empty behavior property, SQL Server Analysis Services trusts that when all of these members
are empty, the calculated member is empty. This means that SQL Server Analysis Services can skip this
calculation. In the expression example above, you could add the [Sale Price] and [Production Cost] to the
Non-empty behavior property, because when they are empty, [Profit] is empty.

Calculated Members in the Calculation Expressions Pane

When you complete the New Calculated Member form, a new CREATE MEMBER MDX command
appears in the Calculation Expressions pane.

The following example shows the code that the Calculated Expressions page displays when you add a
calculated member:

Calculated Member Example

CREATE MEMBER CURRENTCUBE.[Measures].[Profit]
AS ( [Measures].[Sale Price] - [Measures].[Production Cost] ) / [Measures].[Sale
Price]
FORMAT_STRING = "Currency",
NON_EMPTY_BEHAVIOR = { [Measures].[Sale Price], [Measures].[Production Cost] },
VISIBLE = 1 ;
 MCT USE ONLY. STUDENT USE PROHIBITED
5-12 Introduction to MDX

Named Sets
A named set is an MDX expression that returns a
set of dimension members. Named sets can be
defined and saved as part of the cube. Named sets
enable other MDX expressions—for example,
expressions that are issued by clients or custom
applications—to easily access a common set of
data.

Named sets are used for two purposes:

 To enable clients to easily access a commonly

used set of members.

 To enable clients to easily access a set of

members with a complex definition.

By defining a named set, you reduce the amount and complexity of MDX code that must be created in
client applications. As for calculated members, you can use the Calculations tab of the Cube Designer to
create named sets.

The following example creates a set named Core Products that includes only products from the category
Bikes.

Creating a Named Set

CREATE SET CURRENTCUBE.[Core Products]
AS [Product Category].[English Product Category Name].&[Bikes] ;

Scoped Assignments
When you use the New Calculated Member
button or the New Named Set button on the
Calculations tab of the Cube Designer, you add
script to the definition of the cube that creates
that member or set, according to the options that
you select. You can also use the New Script
Command button to add any other kind of MDX
script to the definition of the cube.

A common script to add to the cube definition is a

scoped assignment. This is a way to limit an action
to apply to a subcube, instead of the default
scope, which is the entire cube. Use the SCOPE
statement to begin your scoped assignment and define the subcube. All subsequent operations in your
script will apply to that subcube only until the END SCOPE statement, or until another SCOPE statement.

You can nest SCOPE statements. For example, you could scope the script to provide only results from
2014, take some actions, and then scope the script further to provide only results from January 2014, and
then take some more specific actions.

Within the scoped assignment, use the THIS function to refer to all members of the current subcube.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 5-13

In the following example, a SCOPE statement limits the current scope to the fiscal quarters in 2016 and
the Sales Amount Quota measure. The THIS statement is used to apply a calculation to the members of
that scope by multiplying the quotas from the previous year, which are obtained by using the
PARALLELPERIOD function.

Using the SCOPE Statement

SCOPE
(
[Date].[Fiscal Year].&[2016],
[Date].[Fiscal].[Fiscal Quarter].Members,
[Measures].[Sales Amount Quota]
) ;

THIS = PARALLELPERIOD
(
[Date].[Fiscal].[Fiscal Year], 1,
[Date].[Fiscal].CurrentMember
) * 1.1 ;

END SCOPE

Note: When you are troubleshooting scoped assignments, it is helpful to see which cells are
in the current scope. You can use the BACK_COLOR cell property to highlight the cells in the
current scope.

MDX Functions for Family Members

As you have seen, in an OLAP cube, the members
of dimensions and measures are organized into
hierarchies. When you write MDX expressions and
scripts, it is often helpful to access members that
are related to the current member. For example,
to assign a sales target to a quarter, you might
obtain the sales target for the year and then divide
it by four. The year is the parent of the quarter.

The following functions are used to return various

family members according to their relationship to
the current member:

 PARENT. This function returns the member

one level above the current member. Each member has only one parent. For example, the following
expression returns the parent year for the current quarter.

[Date].[Year].[Quarters].CurrentMember.Parent.Name

 ANCESTOR. This function returns the member a specified number of levels above the current
member. When you specify one level above, ANCESTOR returns the same member as PARENT.
However, you can use ANCESTOR to obtain members multiple levels above the current member in a
single operation. ANCESTOR returns a single member. For example, the following expression returns
the year for the current month.

Ancestor([Date].[Year].[Quarters].[Month].CurrentMember, 2).Name
 MCT USE ONLY. STUDENT USE PROHIBITED
5-14 Introduction to MDX

 CHILDREN. This function returns all of the members, one level below the current member, that have
the current member as their parent.

 FIRSTCHILD. This function returns a single member, one level below the current member, that
appears first in the database.

 LASTCHILD. This function returns a single member, one level below the current member, that
appears last in the database.

 SIBLINGS. This function returns all of the members that have the same parent as the current member.
The members returned include the current member.

 FIRSTSIBLING. This function returns the sibling that appears first in the database.

 LASTSIBLING. This function returns the sibling that appears last in the database.

Other Common MDX Functions

Other MDX functions that are often useful in cube
calculations include the following:

 CURRENTMEMBER. This function returns the

member currently being operated on when
your code iterates through a set of hierarchy
members.

 NAME. This function returns the name of a

cube object. The object might be a dimension,
a hierarchy, a level, or a member.

 DIMENSION. This function returns the

dimension that contains an object. The object
may be a hierarchy, a level, or a member.

 HIERARCHY. This function returns the hierarchy that contains an object. The object might be a level
or a member.
 LEVEL. This function returns the level that contains a member.

 PARALLELPERIOD. This time-based function returns a member from the previous period with the
same relative expression. For example, you could use this function to find a month one year (or four
quarters) earlier than another member, as in the following example:

PARALLELPERIOD([Date].[Calendar].[Calendar Quarter], 4,
[Date].[Calendar].[Month].[October 2016])

 NONEMPTY. This function returns only the members of a set that are not empty. This can help to
improve performance by ensuring that a complex calculation is not performed when a key member is
empty. For example, the following code returns a set of all of the customers whose Sales Amount
value was not empty on January 1, 2016:

NONEMPTY([Customer].Members,
{ [Date].[Calendar].[Date].&[20160101], [Measures].[Sales Amount] })
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 5-15

Demonstration: Adding a Calculated Member to a Cube

In this demonstration, you will see how to add a simple calculated member and then use it in an MDX
query.

Demonstration Steps
Add a Calculated Member

1. Ensure that you have completed the previous demonstration in this module.

2. On the Start screen, type SQL Server Data Tools 2015, and then press Enter.
3. In SQL Server Data Tools, on the File menu, point to Open, and then click Project/Solution.

4. In the Open Project dialog box, browse to the D:\Demofiles\Mod05\Starter folder, and then
double-click Adventure Works OLAP.sln.

5. In Solution Explorer, under Cubes, double-click Sales.cube.

6. In the Cube Designer, click the Calculations tab.

7. Right-click in the Script Organizer pane, and then click New Calculated Member.
8. In the Name text box, type [Total Sales Count].

9. In the Parent hierarchy list, ensure that Measures is selected.

10. In the Expression text box, type the following MDX expression:

[Measures].[Internet Sales Count] + [Measures].[Reseller Sales Count]

11. In the Format string drop-down list, select "0".

12. In the Non-empty behavior list, select Internet Sales Count and Reseller Sales Count, and then
click OK.
13. On the File menu, click Save All.

14. On the Build menu, click Deploy Solution. If you are prompted to authenticate, use the password
Pa55w.rd, and then click OK.
15. When the deployment is complete, close Visual Studio.

Query a Calculated Member

1. Start SQL Server Management Studio.

2. In the Connect to Server dialog box, in the Server type drop-down list, click Analysis Services.

3. In the Server name drop-down list, click MIA-SQL, and then click Connect.
4. In Object Explorer, expand Databases, right-click Adventure Works OLAP, click New Query, and
then click MDX.

5. Type the following code in the query pane:

SELECT
{
[Measures].[Internet Sales Count],
[Measures].[Reseller Sales Count],
[Measures].[Total Sales Count]
} ON COLUMNS,
[Product].[Categorized Products].[Category] ON ROWS
FROM [Sales]

6. Execute the query, and then examine the results.

 MCT USE ONLY. STUDENT USE PROHIBITED
5-16 Introduction to MDX

7. Close SQL Server Management Studio, without saving any changes.

Categorize Activity
Categorize each item into the appropriate category. Indicate your answer by writing the category number
to the right of each item.

Items

1 PARENT

2 PARALLELPERIOD

3 CHILDREN

4 NAME

5 SIBLINGS

6 CURRENTMEMBER

7 FIRSTSIBLING

8 HIERARCHY

9 LASTCHILD

10 LEVEL

11 ANCESTOR

12 NONEMPTY

Category 1 Category 2

Family MDX Non-family

functions MDX
functions
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 5-17

Lesson 3
Using MDX to Query a Cube
Multidimensional data, such as that in an OLAP cube, is often accessed and displayed from Excel in
PivotTables and PivotCharts. It might also be a data source in SQL Server Reporting Services. In this lesson,
you will see how to connect to a cube in SQL Server Analysis Services from these and other tools. After
you have connected to SQL Server Analysis Services, you can issue MDX queries by writing them manually,
or by using user-friendly tools.

Lesson Objectives
After completing this lesson, you will be able to:

 Create connection strings to connect to a specific cube in SQL Server Analysis Services.

 Create MDX queries by using the tools in Excel and SQL Server Reporting Services.

 Advise developers about object models to use for access to SQL Server Analysis Services.
 Query a SQL Server Analysis Services cube from Excel.

Using and Testing MDX Queries

MDX is a query language that enables users to
obtain interesting data quickly from an OLAP
cube. Like Transact-SQL, a typical user has little or
no expertise in MDX, but uses a client application
that has an easily understood user interface to
formulate queries and view results. Typical
scenarios in which MDX is used include the
following:

 From Excel. Excel includes an MDX query

designer that can be used to populate
PivotTables, PivotCharts, and other displays.
 From SQL Server Reporting Services. You
can use Report Builder or Report Designer to build reports that are based on MDX queries, just as you
can for Transact-SQL queries.

 From PerformancePoint Dashboards. In Microsoft SharePoint® sites, you can use

PerformancePoint dashboards to display business performance metrics and compare them against
targets. OLAP cubes can be used as sources for these metrics.

 From a custom application. A developer can write code that uses programmatic interfaces such as
XMLA or ADOMD.NET to send MDX queries to SQL Server Analysis Services. Such code can include
desktop applications, mobile device apps, and many other kinds of custom solution.

Excel, SQL Server Reporting Services, and custom applications will be examined in more detail later in this
module.

Connection and Authentication

Regardless of the client that you use to connect to SQL Server Analysis Services, you must always connect
and authenticate. To connect, you must provide several values to populate a connection string. For SQL
Server Analysis Services, the connection string often includes:
 MCT USE ONLY. STUDENT USE PROHIBITED
5-18 Introduction to MDX

 A data source. This specifies the instance of SQL Server Analysis Services. You can use an instance
name, a fully qualified domain name, an IP address, or a port number to identify the instance.

 An initial catalog. This specifies the SQL Server Analysis Services database to connect to.

 A provider. This optional value specifies the version of SQL Server Analysis Services.

 A cube. This specifies the cube to connect to.

 An application name. By using an application name in your connection string, you identify the client
application. This application name appears in trace logs and can help to identify a problem
application in troubleshooting.
 A timeout. This specifies how long the client waits for a response to a query before it raises an error.

 Character encoding. This specifies whether strings use UTF-8 or UTF-16 to encode characters in
MDX queries.

 An MDX compatibility value. If this value is set to 0 or 1, missing members in ragged hierarchies are
displayed by placeholders. For example, Excel sets this property to 1 and displays an empty cell for
missing members. If this value is set to 2, placeholders are not returned.
SQL Server Analysis Services does not support SQL Server Authentication, so you must use Windows®
integrated authentication. Depending on the client, this may be Kerberos or NTLM authentication. You
can use the following connection string properties to control authentication:
 EffectiveUserName. Use this setting to impersonate a different user for the duration of the
connection. Specify the user account in domain\username format. Note that the connection is made
as the currently logged-on user, who must have administrative permission for SQL Server Analysis
Services.

 Use Encryption for Data. When this value is true, all queries and results are encrypted on the
network between the client and the server.

 User ID. Use this account to connect to SQL Server Analysis Services. If this value is not specified, the
client uses the currently logged-on Windows user account. This setting must be used with Password.

 Password. Use this to authenticate with the User ID setting.

Using SQL Server Management Studio to Test MDX Queries

SQL Server Management Studio is not designed as an end-user tool, but you can use it to connect to SQL
Server Analysis Services and issue MDX queries. Often, SQL Server Management Studio is used to develop
and test queries that are used in reports or PivotTables when you are confident that they execute and
perform as expected.

To connect to SQL Server Analysis Services in SQL Server Management Studio:

1. Start SQL Server Management Studio.

2. In the Connect to Server dialog box, in the Server type drop-down list, select Analysis Services.

3. On the Login tab, in the Server name text box, type the name or fully qualified domain name of the
SQL Server Analysis Services server. To connect to an instance other than the default instance, use the
Server\Instance format.

4. Click Connect.

Note: You can only use Windows authentication to connect to SQL Server Analysis Services,
not SQL Server Authentication. Your current Windows account must be authorized to access the
database and cube.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 5-19

Using MDX from Excel and SQL Server Reporting Services

PivotTables and PivotCharts in Excel are ideal tools
to analyze and display multidimensional data.
Excel can connect to SQL Server Analysis Services
and includes an MDX Query Designer that you can
use to determine what OLAP cube data to import
into Excel.

Connecting to SQL Server Analysis

Services from Excel
Take the following steps to connect to SQL Server
Analysis Services from Excel:

1. Open an existing or new Excel workbook.

2. On the ribbon, click the Data tab.

3. Under Get External Data, click From Other Sources, and then click From Analysis Services.

4. In the Data Connection Wizard, in the Server name text box, type the name of the Analysis Services
server to connect to, and then click Next.
5. From the drop-down list, select the SQL Server Analysis Services database.

6. Select the cube from the list, and then click Finish.

Note: The precise steps for this procedure might vary depending on your version of Excel.

For more information about connecting Excel to SQL Server Analysis Services, see Connect to or import
data from SQL Server Analysis Services:
Connect to or import data from SQL Server Analysis Services
http://aka.ms/Bnxvey

Designing an MDX Query

When you have a connection to SQL Server Analysis Services, you can use the MDX Query Designer to
import data from the cube into Excel. The MDX Query Designer can be used in two modes:

 Design mode. This mode is for those who prefer to construct MDX queries by graphical means. The
user interface includes the following elements:

o Select Cube button. Use this button to select the current cube.

o Metadata pane. This pane displays dimensions, measures, and key performance indicators (KPIs)
in the current cube. You can drag members from this pane onto the data pane to formulate a
query.

o Calculated Members pane. This pane displays calculated members in the current cube. As for
the Metadata pane, you can drag members from this pane onto the data pane.
o Filter pane. Use this pane to create filters from dimensions and measures to limit the data that is
returned.

o Data pane. This is where the results for the current query are displayed.
 MCT USE ONLY. STUDENT USE PROHIBITED
5-20 Introduction to MDX

 Query mode. This mode is for those who prefer to construct queries by writing MDX text. The user
interface includes the following elements:

o Select Cube button. This is the same button that appears in Design mode.

o Metadata pane. As for Design mode, this pane displays dimensions, measures, and KPIs. In
Query mode, however, it also displays MDX functions and template queries, which you can drag
onto the query pane.

o Query pane. This pane displays the current query as MDX query text. You can formulate queries
by typing them manually into this pane, by dragging elements from the Metadata pane, or by a
combination of both operations.

o Data pane. This pane displays the results for the current query as it is displayed in the query
pane.

For more information on how to use the MDX Query Designer in Excel, see Analysis Services MDX Query
Designer (Power Pivot):

Analysis Services MDX Query Designer (Power Pivot)

http://aka.ms/Qaexrf

Using MDX in SQL Server Reporting Services

In SQL Server Reporting Services, you can use an OLAP cube that is hosted in SQL Server Analysis Services
as a source of data to display in a report. To do this, you must create a data source in SQL Server
Reporting Services of the type Microsoft SQL Server Analysis Services.
In SQL Server Reporting Services, data sources can either be embedded, which means that they can only
be used in a single report, or shared, which means that they can be used in many different reports. The
following steps explain how to create a shared data source in SQL Server Analysis Services by using Report
Manager:

1. In Report Manager, on the Contents page, click New Data Source.

2. Type a descriptive name for the data source.

3. In the Data source type drop-down list, select Microsoft SQL Server Analysis Services.

4. In the Connection string text box, type a connection string for the instance of SQL Server Analysis
Services that you want to query. See the previous topic for details of the properties that you should
use.

5. Under Connect using, specify how SQL Server Reporting Services will authenticate with SQL Server
Analysis Services. Remember that you can only use Windows authentication:

o If you select Windows Integrated Security, the current Windows account will be used to
authenticate.

o If you select Credentials supplied by the user running the report, the user will be prompted
for the credentials of a Windows account.

o If you select Credentials stored securely in the report server, the same user account will
always be used to authenticate with SQL Server Analysis Services, regardless of the user who
accesses the report.

o If you select Credentials are not required, SQL Server Reporting Services will authenticate
anonymously with SQL Server Analysis Services. You must ensure that anonymous access is
enabled in SQL Server Analysis Services and also that the unattended execution account is
configured on the report server.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 5-21

6. Use the Test Connection button to validate your settings.

7. Click OK to create the shared data source.

When you have created a data source, you can use the MDX Query Designer to formulate queries. This
tool is very similar to the Excel MDX Query Designer.
For more information about creating MDX queries against SQL Server Analysis Services data in SQL Server
Reporting Services, see Analysis Services Connection Type for MDX (SSRS) in Microsoft Docs:

Analysis Services Connection Type for MDX (SSRS)

http://aka.ms/K3v2bm

Using MDX from Custom Code

SQL Server Reporting Services and Excel are
versatile tools that you can use to display data
from OLAP cubes in many scenarios. However,
sometimes you might want to issue MDX queries
and display results in your own custom
application. Developers have a choice of the
following programmatic interfaces, all of which
can be called from managed code:

 ADOMD.NET. This is the multidimensional

version of the popular ADO.NET object model.
It makes it easy for you to issue MDX queries
and process multidimensional results in
managed code.

 SQL Server Analysis Services OLE DB provider. If you prefer to use the OLE DB standard to access
data, you can use this provider to access a cube.
 XMLA. This is an industry-standard protocol based on SOAP and XML—a lower-level protocol that
might require more custom code than ADOMD.NET. However, it is supported by any programming
platform that has support for SOAP and XML.

 AMO. Analysis Management Objects (AMO) is a library of managed objects that you can use to
configure SQL Server Analysis Services. You cannot use this library to query a cube—instead, you can
use it to create cubes, dimensions, measures, and other objects from custom code.
For more information about programmatic interfaces for multidimensional data, see Data Access with
MDX in Microsoft Docs:

Programmatic Interfaces
http://aka.ms/Cr0cp5
 MCT USE ONLY. STUDENT USE PROHIBITED
5-22 Introduction to MDX

Demonstration: Querying a Cube from Excel

In this demonstration, you will see how to use Excel to import data from a SQL Server Analysis Services
cube into a PivotTable, by formulating an MDX query.

Demonstration Steps
Connect to SQL Server Analysis Services

1. On the Start screen, type Excel 2016, and then press Enter.

2. On the Welcome screen, click Blank workbook.

3. On the ribbon, click the Data tab.

4. Click Get External Data, click From Other Sources, and then click From Analysis Services.

5. In the Server name text box, type MIA-SQL.adventureworks.msft.

6. Select Use the following User name and Password, type the following, and the click Next:

a. User Name: adventureworks\Student

b. Password: Pa55w.rd
7. Ensure that the Adventure Works OLAP database and the Sales cube are selected, and then click
Next.

8. Click Finish.

9. In the Import Data dialog box, ensure that PivotTable Report is selected, and then click OK.

Formulate an MDX Query in Excel

1. In the PivotTable Fields box, under Internet Sales, select Internet Sales Count.
2. Under Reseller Sales, select Reseller Sales Count.

3. Under Values, select Total Sales Count.

4. Under Order Date, select Order Date.Calendar Date.

5. In the COLUMNS box, drag Order Date.Calendar Date to the ROWS box.

6. Examine the data in the PivotTable.

7. Close all open windows, without saving any changes.

 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 5-23

Check Your Knowledge

Question

You are creating a connection string for a

data source in SQL Server Reporting
Services. You want to ensure that ragged
hierarchies include placeholders for
missing members. Which of the following
properties should you use for this
setting?

Select the correct answer.

Data source

Initial catalog

Provider

Character encoding

MDX compatibility
 MCT USE ONLY. STUDENT USE PROHIBITED
5-24 Introduction to MDX

Lab: Using MDX

Scenario
Currently, the AdventureWorks SQL Server Analysis Services cubes contain large amounts of vital
information. However, some of that data needs further calculation to produce the exact information
required. You plan to use MDX calculations to produce this information.

You want to add several important measures to a cube, but these measures must be calculated from
other, existing measures. These include:

 The total revenue from both reseller sales and Internet sales.

 The percentage profit from reseller sales.

 The percentage profit from Internet sales.

You intend to add these measures to the cube by calculating them from existing measures. You also want
to test the cube by using MDX queries in SQL Server Management Studio. Finally, you want to create an
Excel worksheet that draws data from the Sales cube into a worksheet and into a PivotTable.

Objectives
After completing this lab, you will be able to:

 Use form view and script view in the Cube Designer to add calculated members.
 Write and execute MDX queries in SQL Server Management Studio and Excel.

Estimated Time: 45 minutes

Virtual machine: 20768C-MIA-SQL

User name: ADVENTUREWORKS\Student

Password: Pa55w.rd

Exercise 1: Adding Calculated Members

Scenario
You are refining an OLAP cube that summarizes sales data from your data warehouse. You have been
asked to add important calculated members to the measures in the cube. You will add a Total Revenue
member and two profit members—one for Internet sales and one for reseller sales.

The main tasks for this exercise are as follows:

1. Prepare the Lab Environment

2. Add a Calculated Member by Using Form View

3. Add Calculated Members by Using Script View

 Task 1: Prepare the Lab Environment

1. Ensure that the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and then
log on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd.

2. Run Setup.cmd in the D:\Labfiles\Lab05\Starter folder as Administrator.

 Task 2: Add a Calculated Member by Using Form View

1. Start SQL Server Data Tools, and then open the Adventure Works OLAP.sln solution in the
D:\Labfiles\Lab05\Starter folder.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 5-25

2. In the Sales.cube cube, use the Calculations tab of the Cube Designer to add a new calculated
member. Use the following information:

o Name: [Total Revenue]

o Parent hierarchy: Measures

o Expression: Add [Internet Revenue] to [Reseller Revenue]

o Format string: "Currency"

o Non-empty Behavior: [Internet Revenue] and [Reseller Revenue]

 Task 3: Add Calculated Members by Using Script View

1. On the Calculations tab of the Cube Designer, in script view, type MDX code to create a new
calculated member. Use the following information:

o Name: [Internet Profit]

o Expression: Take [Internet Cost] away from [Internet Revenue] and then divide the result by
[Internet Cost]

o Format string: "Percent"

o Non-empty behavior: [Internet Revenue] and [Internet Cost]

o Visible: 1

o Associated Measure Group: 'Internet Sales'

2. Type MDX code to create a second new calculated member. Use the following information:

o Name: [Reseller Profit]

o Expression: Take [Reseller Cost] away from [Reseller Revenue] and then divide the result by
[Reseller Cost]

o Format string: "Percent"

o Non-empty behavior: [Reseller Revenue] and [Reseller Cost]

o Visible: 1

o Associated Measure Group: 'Reseller Sales'

3. Save your changes, deploy the solution, and then close Visual Studio.

Results: After this exercise, you should have added three calculated members to the Sales cube and
deployed the cube to the MIA-SQL instance of SQL Server Analysis Services.
 MCT USE ONLY. STUDENT USE PROHIBITED
5-26 Introduction to MDX

Exercise 2: Querying a Cube by Using MDX

Scenario
You want to test the new calculated members that you added to the cube by using SQL Server
Management Studio to issue MDX queries. You also want to create an Excel workbook that imports data
from the Sales cube into both a worksheet and a PivotTable.

The main tasks for this exercise are as follows:

1. Test Total Revenue in SQL Server Management Studio

2. Test Profit Calculations in SQL Server Management Studio

3. Use Excel to Execute MDX Queries

4. Use an MDX Query in a PivotTable

 Task 1: Test Total Revenue in SQL Server Management Studio

1. Start SQL Server Management Studio, and then connect to the MIA-SQL instance of SQL Server
Analysis Services.

2. Write an MDX query against the [Sales] cube. Return [Internet Revenue], [Reseller Revenue], and
[Total Revenue] on columns, and all of the members of [Product].[Categorized Products] on rows.
Examine the results to ensure that the total is correct for each member.

 Task 2: Test Profit Calculations in SQL Server Management Studio

1. In SQL Server Management Studio, create and execute a new MDX query against the Adventure
Works OLAP cube. Use the following information:

o Columns: [Internet Cost], [Internet Revenue], and [Internet Profit]

o Rows: All of the members of [Customer].[City]

2. Add a slicer axis to the MDX query so that it displays only results from New York State. Execute your
modified query, and then examine the results.

3. Create and execute a second new query. Use the following information:
o Columns: [Reseller Cost], [Reseller Revenue], and [Reseller Profit]

o Rows: All of the members of [Sales Territory].[Sales Territory]

4. Close SQL Server Management Studio, without saving any changes.

 Task 3: Use Excel to Execute MDX Queries

 In Excel 2016, add a new SQL Server Analysis Services query that loads data from the Sales cube in
the Adventure Works OLAP database. Display the following columns:

o Internet Cost

o Internet Revenue

o Internet Profit

o Product Category

o Product Subcategory
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 5-27

 Task 4: Use an MDX Query in a PivotTable

1. In Excel, use the Get External Data tool to connect to SQL Server Analysis Services to import sales data
into a new PivotTable. Use the following information:

o Server name: MIA-SQL

o Database: Adventure Works OLAP

o Cube: Sales
o Import the data into a new worksheet

2. In the new PivotTable, choose the Internet Revenue, Reseller Revenue, and Total Revenue
measures for columns, and the Sales Territory dimension for rows.

3. In the worksheet, determine the Total Revenue value for the United Kingdom.

4. Close Excel, without saving any changes.

Results: After this exercise, you should have created several test MDX queries in SQL Server Management
Studio and an Excel spreadsheet that connects to the Sales OLAP cube.
 MCT USE ONLY. STUDENT USE PROHIBITED
5-28 Introduction to MDX

Module Review and Takeaways

MDX is a sophisticated query language that is much like Transact-SQL, but designed specifically for
executing queries against, and returning, multidimensional data such as that in an OLAP cube. In this
module, you have seen how to specify query axes by using the SELECT statement, how to specify a slicer
axis by using the WHERE clause, and how to establish cube context by using the FROM clause. You have
learnt how advanced features such as calculated members and named sets can be added to a cube by
using MDX queries in the Cube Designer. You have also seen how to display cube data in client
applications such as SQL Server Reporting Services and Excel.

Review Question(s)
Question: Users often want to aggregate data from Europe and North America in their
reports and spreadsheets. Users who have good MDX skills have been able to do this by
editing MDX queries in their client programs. How can you help users who have no MDX
skills to display this data?

Question: You are testing the Sales cube by writing MDX queries in SQL Server
Management Studio. You would like to compare sales for 2015 with data from the Archived
Sales cube, which includes sales data from 2008 to 2010. How can you add this data to your
query?
 MCT USE ONLY. STUDENT USE PROHIBITED
6-1

Module 6
Customizing Cube Functionality
Contents:
Module Overview 6-1 
Lesson 1: Implementing Key Performance Indicators 6-2 

Lesson 2: Implementing Actions 6-8 

Lesson 3: Implementing Perspectives 6-11 

Lesson 4: Implementing Translations 6-14 

Lab: Customizing a Cube 6-16 

Module Review and Takeaways 6-20 

Module Overview
In this module, you will learn how to customize cube functionality by using several technologies that are
available in SQL Server Analysis Services. By customizing cubes, you can improve the user experience,
making cubes easier to use.
You will learn about key performance indicators (KPIs) and how they can be used to obtain a quick and
accurate summary of business progress. You will find out about actions, and how they enable end users to
go beyond traditional analysis to initiate solutions, and to discover problems and deficiencies. This
module also covers perspectives and how they enable users to see a simplified version of a cube, so that
users can focus on the most relevant data. Finally, you will learn about translations, and how they can be
used to translate various elements of a cube, and enable users to view and understand cube and
dimension data.

Objectives
After completing this module, you will be able to:

 Implement KPIs

 Implement actions

 Implement perspectives

 Implement translations
 MCT USE ONLY. STUDENT USE PROHIBITED
6-2 Customizing Cube Functionality

Lesson 1
Implementing Key Performance Indicators
In business, a KPI is a quantifiable measurement that is used to understand how well, or badly, an activity
is going. KPIs are mostly evaluated over time. KPIs may be specific to a department, or generic for the
whole organization.

Generic KPI examples might include gross profit or sales turnover measured monthly. Department-specific
KPIs might include staff turnover, and the number of people who have been trained in a month for the
human resources department.

KPIs are often viewed together in a business scorecard to give a summary of the health of a business.

Lesson Objectives
After completing this lesson, you will be able to:

 Describe KPIs.
 List the elements of a KPI.

 Browse a KPI.

Introducing KPIs
In SQL Server Analysis Services, a KPI is a group of
calculations that are used to evaluate a business
activity. KPIs are associated with a measure group
in a cube, and can be compared against a business
goal to provide trend information over time.

A multidimensional expression in the

Multidimensional Expressions (MDX) language can
be used to return a KPI status value between -1
(worst) and +1 (best), which is then passed to a
client application. A typical graphical
representation might be to create a performance
dashboard that has traffic light indicators. In this
example, a value between -1.0 and -0.34 might display a red traffic light, a value between -0.33 and +0.33
an amber traffic light, and a value between +0.34 and +1.0 a green traffic light. As you will see later in the
module, a KPI status value is a value compared against a goal. You monitor KPI status values over time
and they provide a trend.
In practice, business rules will determine how to assign the KPI status values. An unsuitable MDX formula
will result in a department, or the company, receiving incorrect feedback on their progress.

Best Practice: Getting a KPI measurement right is a business decision rather than an IT
decision. A KPI must genuinely reflect an outcome that the business wants to achieve.
Care must be taken that the KPI does not encourage damaging behavior. For example,
measuring sales turnover by quarter could result in salespeople booking work before client
contracts have been signed, or agreeing sales at reduced margins. A KPI that measures repeat
purchases, on the other hand, might be a better indicator of customer satisfaction and future
business success.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 6-3

An advantage of calculating KPIs by using SQL Server Analysis Services is that the data is held centrally,
and can be used by different client applications. When KPIs are calculated locally, they might be
calculated by using different datasets, and could be difficult to compare.

SQL Server Analysis Services includes functionality that helps you to define KPI calculations in a
multidimensional cube.
For more information about KPIs, see Key Performance Indicators (KPIs) in Multidimensional Models in
Microsoft Docs:

Key Performance Indicators (KPIs) in Multidimensional Models

http://aka.ms/Fs3xyp

Elements of a KPI
KPIs in SQL Server Analysis Services consist of
several properties, including:

 Value

 Goal

 Status, and status indicator

 Trend, and trend indicator

 Display folder

 Parent KPI

 Weight

 Current time member

Value
A value is the result of a calculation that is defined by using an MDX expression. This is the actual value of
the KPI and represents the current position.

Goal
A goal expression is a value, or an MDX expression that resolves to a value, that defines the target for the
measure that the value expression describes. For example, a retailer might have a margin goal expression
that has a value of 0.6 (or 60 percent) for accessories, and 0.3 (30 percent) for all other categories.

Margin goal of 60 percent for accessories, and 30 percent for all other categories.

MDX Expression
Case
When [Product].[Category].CurrentMember Is
[Product].[Category].[Accessories]
Then .60
Else .30
End

Status
A status expression evaluates the current status of the KPI value, compared to the goal. The result is
returned within the range of -1 to +1, where -1 is the worst and +1 is the best. The following MDX
expression returns a value of 1 if the KPI value is greater than, or equal to, the goal value. It returns a
 MCT USE ONLY. STUDENT USE PROHIBITED
6-4 Customizing Cube Functionality

value of -1 if the KPI value is less than 0.5; otherwise, it returns 0. This means that the status is 1 if the goal
has been reached, 0 if the value is more than halfway toward the goal, and -1 if the value is less than
halfway toward the goal.

The KPI value is compared to the Goal value to return a Status value.

MDX Expression
Case
When KpiValue( "Gross Profit Margin" ) /
KpiGoal ( "Gross Profit Margin" ) >= 1
Then 1
When KpiValue( "Gross Profit Margin" ) /
KpiGoal ( "Gross Profit Margin" ) < 0.5
Then -1

Else 0
End

Trend
A trend expression is used to evaluate the current trend of the value expression compared to the goal
expression. The trend expression helps a business user to understand whether a KPI value is getting better
or worse over time, relative to the goal expression.

You can also associate graphical devices with a trend expression to help business users to quickly
understand the trend.
The following code returns a value of 1 if the gross profit margin is increasing over the last year, -1 if it is
decreasing, and 0 if it is the same.

KPI Trend Expression

Case
When KpiValue( "Gross Profit Margin" ) >
(KpiValue( "Gross Profit Margin" ),
ParallelPeriod
( [Date].[Fiscal Time].[Fiscal Year], 1,
[Date].[Fiscal Time].CurrentMember ))
Then 1
When KpiValue( "Gross Profit Margin" ) <
(KpiValue( "Gross Profit Margin" ),
ParallelPeriod
( [Date].[Fiscal Time].[Fiscal Year], 1,
[Date].[Fiscal Time].CurrentMember ))
Then -1
Else 0
End

Additional properties of the KPI include:

 Display folder. This is where the KPI will appear if you are browsing the cube.

 Parent KPI. This property defines the parent of the current KPI. The browser displays the KPI as a
child while allowing the parent to access the values of the child. This feature enables you to have KPIs
that are based on other KPIs.

 Weight. You can apply a weight to adjust the importance of a child KPI against its siblings.

 Current time member. This is an MDX expression that defines the current time member for the KPI.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 6-5

Browsing KPIs
KPIs in SQL Server Analysis Services are server-
based, so they can be used by different client
applications. The business logic is within SQL
Server Analysis Services, and stored in the cube. In
addition, client applications can use MDX
expressions with the cube data to retrieve parts of
the KPI—such as the value or goal—to use in
expressions, statements, and scripts.

Third-party client applications might not display

the indicators that you have defined. You should
verify that the KPI displays correctly.

The following code example uses MDX functions

to return the KPI value and goal for the Reseller Margin KPI of the Adventure Works database for
descendants of three members of the Calendar Year attribute hierarchy.

Reseller profit and goal returned by using MDX functions.

Reseller Margin KPI

SELECT
{
KPIValue("Reseller Margin"),
KPIGoal("Reseller Margin")
} ON COLUMNS,
{
[Order Date].[Calendar Date].[Calendar Year].&[2006],
[Order Date].[Calendar Date].[Calendar Year].&[2007],
[Order Date].[Calendar Date].[Calendar Year].&[2008]
} ON ROWS
FROM [Sales];

Demonstration: Creating a KPI

In this demonstration, you will see how to:
 Create a KPI.

 Browse a KPI in the Cube Designer.

 Browse a KPI in Excel.

Demonstration Steps
Create a KPI
1. Ensure that the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and then
log on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd.

2. In the D:\Demofiles\Mod06 folder, run Setup.cmd as Administrator.

3. In the User Account Control dialog box, click Yes.

4. Start Microsoft® Visual Studio®, and then in the D:\Demofiles\Mod06 folder, open the
Adventure Works OLAP.sln solution.
 MCT USE ONLY. STUDENT USE PROHIBITED
6-6 Customizing Cube Functionality

5. On the Build menu, click Deploy Solution.

6. If the Microsoft Visual Studio dialog box appears, click Yes.

7. If the Account Password dialog box appears, enter the password Pa55w.rd for
ADVENTUREWORKS\ServiceAcct, and then click OK.

8. After deployment is complete, close the Deployment Progress window.

9. In Solution Explorer, double-click Sales.cube to open it in the Cube Designer, and then click the
Calculations tab.

10. In the Script Organizer pane, select [Reseller Profit], and then note that this calculated member
calculates reseller gross margin as a percentage of revenue.

11. In the Cube Designer, click the KPIs tab, and then, on the Cube menu, click New KPI.
12. In the Name box, type Reseller Margin, and then in the Associated measure group list, select
Reseller Sales.

13. In the Calculation Tools pane, on the Metadata tab, expand Measures, expand Reseller Sales, and
then drag the Reseller Profit measure to the Value Expression box.

14. In the Goal Expression box, type 0.1 (this sets a gross margin target of 10 percent).

15. Check that Gauge is selected in the Status indicator list, and then type the following MDX
expression in the Status expression box (you can copy this from New_KPI_Demo.txt in the demo
folder):

Case
When
KpiValue("Reseller Profit")/KpiGoal("Reseller Profit")>=.95
Then 1
When
KpiValue("Reseller Profit")/KpiGoal("Reseller Profit")<.95
And
KpiValue("Reseller Profit")/KpiGoal("Reseller Profit")>=.5
Then 0
Else -1
End

16. Check that Standard arrow is selected in the Trend indicator list, and then type the following
expression in the Trend expression box (you can copy this from New_KPI_Demo.txt in the demo
folder):

Case
When IsEmpty(ParallelPeriod([Order Date].[Calendar Date].[Calendar Year],
1, [Order Date].[Calendar Date].CurrentMember))
Then 0
When [Measures].[Reseller Profit] = (ParallelPeriod([Order Date].[Calendar
Date].[Calendar Year],
1, [Order Date].[Calendar Date].CurrentMember), [Measures].[Reseller Profit])
Then 0
When [Measures].[Reseller Profit] >
(ParallelPeriod([Order Date].[Calendar Date].[Calendar Year],
1, [Order Date].[Calendar Date].CurrentMember), [Measures].[Reseller Profit])
Then 1
Else -1
End

17. On the File menu, click Save All.

 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 6-7

Browse a KPI in the Cube Designer

1. On the Build menu, click Deploy Solution. If you are prompted for impersonation credentials, enter
the password Pa55w.rd for ADVENTUREWORKS\ServiceAcct.

2. After deployment is complete, close the Deployment Progress window.

3. In the cube browser, click the KPIs tab, and then on the Cube menu, point to Show KPIs in, and then
click Browser. The KPI is shown based on the total overall margin.

4. In the top pane of the KPI browser, in the Dimension list, select Sales Territory, in the Hierarchy list,
select Sales Territory, in the Operator list, select Equal, and then in the Filter Expression list,
expand All, select North America, and then click OK.

5. Click anywhere in the KPI browser pane to update the values for the Reseller Margin KPI. The updated
values indicate the overall margin for sales in North America.

6. Amend the KPI criteria again. In the top pane of the KPI browser, in the Dimension list, select Order
Date, in the Hierarchy list, click Calendar Date, in the Operator list, select Equal, and then in the
Filter Expression list, expand All, select 2006, and then click OK.
7. Click anywhere in the KPI browser pane to update the values for the Reseller Margin KPI. The updated
values indicate the overall Reseller margin for sales in 2006.

Browse a KPI in Excel

1. In the Cube Designer, click the Browser tab, and then on the Cube menu, click Analyze in Excel. If a
security notice is displayed, click Enable.

2. In Microsoft Excel®, in the PivotTable Fields pane, under Reseller Sales, select Reseller Profit. The
overall gross margin is shown in the PivotTable.

3. In the PivotTable Fields pane, under Order Date, drag Order Date.Calendar Date to the Rows
area. The gross margin for each year is shown.
4. In the PivotTable Fields pane, expand KPIs, expand Reseller Margin, and then select Goal. The
Reseller Margin goal is displayed in the PivotTable.

5. Close Excel without saving the workbook.

6. Close Visual Studio.

Question: What KPIs are used in your business? Do you have KPIs in your department? Does
the KPI generally have a positive effect on employee behavior? Can you think of any KPIs
that have had unwanted results?
 MCT USE ONLY. STUDENT USE PROHIBITED
6-8 Customizing Cube Functionality

Lesson 2
Implementing Actions
In SQL Server Analysis Services, an action is a stored MDX statement that can be presented to and
employed by client applications. An action enables SQL Server Analysis Services to interact with other
applications, providing information that can be used to perform a task, such as opening a web page or
displaying data. The action is defined on the server and used by a client application.

Lesson Objectives
After completing this lesson, you will be able to:

 Describe actions.

 List the types of actions.

 Implement actions.

Introducing Actions
An action is an MDX statement that is saved as
part of a cube definition, and can be used by
client applications. Actions are stored on the
server, and perform some action that is required
on the client side. For example, an action might be
to drill through and return a rowset, or to display a
web page in a browser. The action is stored in the
SQL Server Analysis Services database, so it is
important to test whether the client application
supports the action and can correctly interpret the
events of the action.

A simple action object consists of basic

information, the target, a condition to limit the action scope, and the type. The target is the location in
the cube where the action is to occur, and consists of a target type and target object.

The target type could be level members, cells, hierarchy, hierarchy members, or others. The target object
is specific to the target type. For example, if the target type is hierarchy, the target object is any one of the
defined hierarchies in the cube.

A condition is implemented as a Boolean MDX statement that evaluates to true or false. If the statement
returns true, the action is performed; if it returns false, the action is not performed.

By saving and reusing actions, end users can go beyond traditional analysis, which finishes with data
presentation. Actions enable users to look for solutions to issues that they discover as they work with the
data. Actions enable a business intelligence application to go beyond the cube.

You can be flexible when you create actions. For example, an action can launch an application, retrieve
information from a database, or open a web page in a browser.

You can configure actions to be triggered from almost any part of a cube, including dimensions, levels,
members, and cells, or to create multiple actions for the same portion of a cube.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 6-9

Types of Actions
SQL Server Analysis Services supports different
types of actions:

 CommandLine. Executes a command at the

command prompt.

 Dataset. Returns a dataset.

 Drillthrough. Returns an expression that the

client application then uses to return a rowset.

 Html. Executes some HTML script.

 Proprietary. Uses a proprietary interface to
perform an action.

 Report. Returns a report based on a parameterized URL.

 Rowset. Returns a rowset.

 Statement. Runs an OLE DB command.

 URL. Displays a web page.

Typically, actions pass an attribute of the current object, enabling them to be dynamic. For example, you
could create a URL action that displays a customer’s location on a map. To create this action, open
Microsoft Bing® Maps, search for a postal code, and then click Share. Bing will generate a URL that
contains the postal code, which you can then remove and append to the original URL.
For more information about actions in SQL Server Analysis Services, see Actions (Analysis Services -
Multidimensional Data) in Microsoft Docs:

Actions (Analysis Services - Multidimensional Data)

http://aka.ms/d2i6vg

Creating Actions for a Cube

Use the Actions tab of the Cube Designer to build
actions and specify the following details for
standard actions:

1. Name. Select a name that identifies the

action.

2. Action Target. Select the object to which the

action is attached. For Target type, select
from the following objects:

o Attribute members

o Cells

o Cube

o Dimension members

o Hierarchy
 MCT USE ONLY. STUDENT USE PROHIBITED
6-10 Customizing Cube Functionality

o Hierarchy members

o Level

o Level members

3. Action Content. Select the type of action from the following:

o Dataset. This retrieves a dataset.

o Proprietary. This performs an operation by using an interface other than those listed here.

o Rowset. This retrieves a rowset.

o Statement. This runs an OLE DB command.

o URL. This displays a variable page in an Internet browser.

4. Additional Properties. Select additional action properties from the following:

o Invocation. This specifies how the action is run. Interactive, the default, specifies that the action
is run when a user accesses an object. The possible settings are Batch, Interactive, and On Open.

o Application. This describes the application of the action.

o Description. This describes the action.

o Caption. This provides a caption that is displayed for the action. If the caption is generated
dynamically by an MDX statement, specify True for Caption is MDX.
o Caption is MDX. This indicates that the caption is an MDX expression that, after it is executed,
will generate the text to be displayed.

To create a new drill-through action, on the Cube menu, click New Drillthrough Action, and then
specify the drill-through columns that are required.
To create a reporting action, on the Cube menu, click New Reporting Action, and then specify the
Server Name, Server Path, and Report Format values.

Note: For Server Path, specify the path to the report on the instance of SQL Server
Reporting Services. For example, AdventureWorks/YearlyInternetSales. For Report Format, select
HTML5, HTML3, Excel, or PDF based on the client tool that your users are most likely to use to
access your report.

Verify the correctness of the statement by placing a mark in the column to the right.

Statement Answer

True or false? Actions are client-

side processes that return data to
the server.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 6-11

Lesson 3
Implementing Perspectives
Cubes can be complex for users to explore in SQL Server Analysis Services.

A single cube can represent the contents of a complete data warehouse, with multiple measure groups
that represent multiple fact tables, and multiple dimensions based on multiple dimension tables. Although
cubes are powerful, they can be daunting. A user might only need to interact with a small part of the cube
to satisfy their business intelligence and reporting requirements.

In SQL Server Analysis Services, you can use a perspective to reduce the perceived complexity of a cube.

Lesson Objectives
After completing this lesson, you will be able to:

 Describe perspectives.

 Implement perspectives.

Introducing Perspectives
A perspective is a view of a cube, helping users to
focus on a specific aspect of cube data.

Although cubes are more intuitive to navigate

than relational databases, they can become large if
they include many measure groups, measures, and
dimensions. However, perspectives do not cause
any additional processing overhead, so you can
create them for each view of the data that users
need.
A simple perspective object consists of basic
information, dimensions, measure groups,
calculations, KPIs, and actions. Each of these
properties defines a subset of the cube as follows:
 Basic information includes the name and default measure of the perspective.

 The dimensions are a subset of the cube dimensions.

 The measure groups are a subset of the cube measure groups.

 The calculations are a subset of the cube calculations.

 The KPIs are a subset of the cube KPIs.

 The actions are a subset of the cube actions.

Objects in a cube that are not visible to the user through a perspective can still be directly referenced and
retrieved by using XML for Analysis (XMLA), MDX, or Data Mining Extensions (DMX) statements.

Perspectives do not restrict access to objects in a cube and should not be used as such. Instead, use
perspectives to provide a better experience for users as they work with cube data.
 MCT USE ONLY. STUDENT USE PROHIBITED
6-12 Customizing Cube Functionality

For more information about perspectives, see Perspectives in Microsoft Docs:

Perspectives
http://aka.ms/K7c015

Defining Perspectives in the Cube Designer

The first column of the Perspectives tab in the
Cube Designer is Cube Objects, which lists all of
the objects in the cube. You can add a perspective
to the Perspectives tab by:

 Clicking New Perspective on the Cube menu.

 Clicking the New Perspective button on the

toolbar.

 Right-clicking anywhere in the pane, and then

clicking New Perspective on the shortcut
menu.

Naming a Perspective
When you create a perspective, the name defaults to Perspective. Additional perspectives default to
Perspective 1, Perspective 2, and so on. You should change the name to something meaningful that
describes the purpose of the perspective to business users. Then use the checkboxes to define which
measures, dimension attributes and hierarchies, KPIs, named sets, and calculated members to include in
the perspective.

Removing a Perspective
You can remove a perspective by:

 Clicking any cell in the column for the perspective that you want to delete. Then, on the Cube menu,
click Delete Perspective.

 Clicking the Delete Perspective button on the toolbar.

 Right-clicking any cell in the perspective that you want to delete, and then clicking Delete
Perspective on the shortcut menu.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 6-13

Check Your Knowledge

Question

How many perspectives can each cube

contain?

Select the correct answer.

Each cube can contain no more than

one perspective.

Each cube can contain many

perspectives.

Each cube must contain one

perspective for each user.

The number of perspectives depends

on the security rules of the cube.

Cubes do not support perspectives in

SQL Server Analysis Services.
 MCT USE ONLY. STUDENT USE PROHIBITED
6-14 Customizing Cube Functionality

Lesson 4
Implementing Translations
Multilanguage support in SQL Server Analysis Services is accomplished by using language translations. In
this lesson, you will be introduced to the concepts that are involved in using translations, and how to
implement cube and dimension translations.

Lesson Objectives
After completing this lesson, you will be able to:

 Describe translations.

 Implement cube translations.

 Implement dimension translations.

Introducing Translations
A translation is a mechanism to change the
displayed labels and captions from one language
to another.
Translating elements of a cube into a different
language enables more users to view and
understand the cube's metadata. Translations are
available for all objects in SQL Server Analysis
Services, and can also be used to display an
alternative column. Therefore, if you have data in
multiple languages, a different column could be
displayed for an attribute in each language.
A simple translation object is composed of a
language ID number—an integer with the language ID—and a translated caption that is the translated
text.

In SQL Server Analysis Services, a cube translation is a language-specific representation of the name of a
cube object, such as a caption or display folder. There is no automatic translation functionality, so you
must provide text for each label in each language.

Translations provide server support for applications that can support multiple languages. For others, the
default language is displayed.

The collation and language information for the client computer is stored in a locale identifier (LCID). The
following process takes place when a client connects to the cube:

 On connection, the client passes the LCID to the instance of SQL Server Analysis Services.

 The instance uses the LCID to determine which set of translations to use when providing metadata for
objects in SQL Server Analysis Services to each business user.

 If an object in SQL Server Analysis Services does not contain the specified translation, the default
language is used to deliver content to the client.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 6-15

For more information about implementing translations, see Translation support in Analysis Services in
Microsoft Docs:
Translation support in Analysis Services
http://aka.ms/Divre7

Implementing Cube Translations

In a cube, you can specify language translations
for the following objects:

 Measure groups and measures

 Dimensions

 KPIs

 Named sets

 Calculated members
To create a cube translation, click the New
Translation button on the toolbar of the
Translations tab in the Cube Designer. To remove
a translation, select it, and then click Delete Translation.

Implementing Dimension Translations

You can specify translations for the following
objects in a dimension:

 Dimension

 Attributes

 Hierarchies and levels

To create a dimension translation, click the New

Translation button on the toolbar of the
Translations tab in the Dimension Designer. To
remove a translation, select the translation, and
then click Delete Translation.

In addition to specifying translated values for object names, you can also translate attribute members by
specifying another column in the dimension table that contains translated values. Select New Caption
Column to display the Attribute Data Translation dialog box, and then define a new caption column
when you modify an attribute in the Translation Details grid. For example, you could specify a caption
column of French Month Name for the Month attribute of the Time dimension. You can use the Edit
Caption Column and Delete Caption Column buttons to modify or delete caption columns.

Question: What do you think are the advantages and disadvantages of implementing
translations in a cube?
 MCT USE ONLY. STUDENT USE PROHIBITED
6-16 Customizing Cube Functionality

Lab: Customizing a Cube

Scenario
Information workers in the sales department at Adventure Works Cycles use the cube that you have
created to analyze sales data. Several users have asked to be able to view aggregated sales figures in
Excel, and then drill through to see details of specific orders for a given aggregation.

Some users are complaining that the cube is too complex for their needs. These users typically only want
to analyze sales amount by customer, and would like to view the measures and dimensions that are
needed to support this requirement. Finally, the company employs several senior sales managers who
would like to view cube data and metadata in French, their first language.

Objectives
After completing this lab, you will be able to:

 Implement an action

 Implement a perspective
 Implement a translation

Estimated Time: 45 minutes

Virtual machine: 20768C-MIA-SQL

User name: ADVENTUREWORKS\Student

Password: Pa55w.rd

Exercise 1: Implementing an Action

Scenario
Information workers in the sales department at Adventure Works Cycles are using the cube that you have
created to analyze sales data. Several users want to view aggregated sales figures in Excel, and then
quickly drill through to see details of specific orders for a given aggregation. For example, when the users
view the sales total for a product category in a specific year, they need to quickly generate a second Excel
worksheet that shows details of the customer, location, date, and individual product for each sale in that
year.

The main tasks for this exercise are as follows:

1. Prepare the Lab Environment

2. Create a Drill-Through Action

3. Browse a Drill-Through Action

 Task 1: Prepare the Lab Environment

1. Ensure that the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and then
log on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd.

2. Run Setup.cmd in the D:\Labfiles\Lab06\Starter folder as Administrator.

 Task 2: Create a Drill-Through Action

1. Use Visual Studio to open and deploy the Adventure Works OLAP.sln solution in the
D:\Labfiles\Lab06\Starter folder.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 6-17

2. If you are prompted for impersonation credentials, enter the password Pa55w.rd for
ADVENTUREWORKS\ServiceAcct.
3. Open Sales.cube, and then add a drillthrough action named Internet Sales Details.

4. The action should use the Internet Sales measure group members.

5. The action should return the specified drill-through columns from the following dimensions:

a. Customer (City and Full Name)

b. Order Date (Full Date Alternate Key)

c. Product (English Product Name)

6. Set additional properties so that the drill-through action has the caption Drillthrough to Order
Details.

7. Save the solution when you have finished.

 Task 3: Browse a Drill-Through Action

1. Deploy the Adventure Works OLAP solution. If you are prompted, enter the password Pa55w.rd.

2. In the Cube Designer, select the Browser tab, and then on the Cube menu, click Analyze in Excel.

3. Choose PivotTable Fields:

a. Internet Revenue measure

b. Order Date.Calendar Date

c. Product.Categorized Products

4. Right-click the sales amount for Bikes in 2007, point to Additional Actions, and then click
Drillthrough to Order Details to test the drill-through action.

5. When you have finished, close Excel without saving the workbook.

Results: After this exercise, you should have a new drill-through action in the solution.

Exercise 2: Implementing Perspectives

Scenario
Some of the users in the sales department are complaining that the cube is too complex for their needs.
They explain that they typically need to analyze sales amount by customer, and want to view the
measures and dimensions that are required to support this.

The main tasks for this exercise are as follows:

1. Create a Perspective

2. Browse a Perspective

 Task 1: Create a Perspective

1. Add a perspective named Internet Sales to the Sales cube. The perspective should include only cube
objects that are relevant to reseller sales:
1. Remove all measures for Reseller Sales.

2. Remove the Reseller dimension.

 MCT USE ONLY. STUDENT USE PROHIBITED
6-18 Customizing Cube Functionality

3. Remove the Reseller Profit calculated member.

4. Add a perspective named Reseller Sales. The perspective should include only cube objects that are
relevant to reseller sales:

1. Remove the Internet Sales measure group.

2. Remove the Customer dimension.

3. Remove the Internet Sales Details action.

4. Remove the Internet Profit calculated member.

5. Save the solution.

 Task 2: Browse a Perspective

1. In Visual Studio, on the Build menu, click Deploy Solution to deploy the Adventure Works OLAP
solution.

2. Select the Browser tab on the Sales cube.

3. Click the ellipsis (…) above the Metadata pane to select the Internet Sales perspective.

4. In the Metadata pane:

a. Expand Internet Sales, and then drag Internet Revenue to the query results pane.
b. Expand the Customer dimension, and then drag Yearly Income to the query results area.

5. Analyze the cube in Excel to view the Reseller Sales perspective.

6. In Excel, in the PivotTable Fields pane, select Reseller Revenue from Reseller Sales and Business
Type from Reseller.

7. When you have finished browsing the perspectives, close Excel without saving the workbook.

Results: After this exercise, you should have defined a new perspective in the solution.

Exercise 3: Implementing a Translation

Scenario
Adventure Works Cycles employs two French sales managers. They have asked to be able to view cube
data, in addition to metadata, in their native language.
In this lab exercise, you will add the final enhancement to the revenue information cube by specifying
French translations.

The main tasks for this exercise are as follows:

1. Create Dimension Translations

2. Create Cube Translations

3. Browse Translations

 Task 1: Create Dimension Translations

1. Using the dimension designer for the Date dimension, create a French (France) translation. This
should include translated captions for the following items:

a. The Calendar Date hierarchy—translation Date du Calendrier.

 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 6-19

b. The Calendar Year level— translation Année.

c. The Calendar Semester level—translation Semestre.

d. The Calendar Quarter level—translation Trimestre.

e. The Month level—translation Mois.

f. The Day level—translation Journée.

Note: To type é, press Alt+130. Check that Num Lock is on. Otherwise, type the translation
without the accents.

2. Configure the Month attribute to use FrenchMonthName as a translation column.

3. To display hidden attributes, click the Show All Attributes toolbar icon.

4. To configure attribute translation columns, click the France (French) language column for the
attribute. Select FrenchMonthName.
5. Save the Date dimension when you have finished.

 Task 2: Create Cube Translations

1. Using the Cube Designer for the Sales cube, create a new French (France) translation that has the
following caption translations:

a. The Internet Sales measure group—translation Ventes d'Internet.

b. The Internet Revenue measure—translation Revenu d’Internet.

c. The Order Date dimension—translation Date de Vente.

2. Save the cube when you have finished.

 Task 3: Browse Translations

1. Deploy the solution. If you are prompted for impersonation credentials, enter the password Pa55w.rd
for ADVENTUREWORKS\ServiceAcct.

2. Use the Browser tab of the Cube Designer to browse the cube, reconnecting if necessary.

3. In the Language list, select French (France).

4. View the Revenu d’Internet measure by the Date de Vente.Date du Calendrier hierarchy.

5. Verify that the French translations are used for the captions and month attribute values.

6. Close Visual Studio.

Results: After this exercise, you should have French data and metadata in the cube.

Question: When you apply the translation feature to a cube, what appears in the specified
language?

Question: What does a drill-through action do?

 MCT USE ONLY. STUDENT USE PROHIBITED
6-20 Customizing Cube Functionality

Module Review and Takeaways

In this module, you have learned how to use KPIs, actions, perspectives, and translations. These features
are designed to help you to make it easier for users to work with cube data.

Review Question(s)
Question: Can you think of scenarios in your own business where the use of the cube
enhancements discussed in this module would help business users?
 MCT USE ONLY. STUDENT USE PROHIBITED
7-1

Module 7
Implementing a Tabular Data Model by Using SQL Server
Analysis Services
Contents:
Module Overview 7-1 

Lesson 1: Introduction to Tabular Data Models in SQL Server Analysis Services 7-2 

Lesson 2: Creating a Tabular Data Model 7-7 

Lesson 3: Using a SQL Server Analysis Services Tabular Model in an Enterprise
BI Solution 7-18 

Lab: Implementing a Tabular Data Model in SQL Server Analysis Services 7-26 

Module Review and Takeaways 7-36 

Module Overview
You can install SQL Server® Analysis Services in tabular mode and create tabular data models that
information workers can access by using tools such as Microsoft® Excel® and Power View.

This module describes tabular data models in SQL Server Analysis Services and explains how to develop a
tabular data model by using SQL Server Data Tools.

Objectives
After completing this module, you will be able to:
 Describe considerations for tabular databases.

 Create a tabular data model.

 Use a tabular model in SQL Server Analysis Services.

 MCT USE ONLY. STUDENT USE PROHIBITED
7-2 Implementing a Tabular Data Model by Using SQL Server Analysis Services

Lesson 1
Introduction to Tabular Data Models in SQL Server
Analysis Services
This lesson describes tabular data models in SQL Server Analysis Services and explains how you can use
SQL Server Data Tools to create a tabular data model.

Lesson Objectives
After completing this lesson, you will be able to:

 Explain tabular data models in SQL Server Analysis Services.

 Describe the options for creating a tabular data model in SQL Server Analysis Services.

 Explain the benefits of using SQL Server Data Tools to develop a tabular data model.

 Describe the purpose of the workspace database.

What Is a Tabular Data Model?

Tabular data models are in-memory databases
that use the xVelocity storage and processing
engine. The xVelocity engine uses column-based
storage and sophisticated compression algorithms
to deliver very fast query response times, even
with large datasets that contain millions of rows of
data. This makes the xVelocity engine ideal for the
complex queries that data analysis and reporting
tools typically generate.

Tabular data models expose data in a relational

format, so the developer interacts with tables and
relationships instead of dimensions and cubes.
Tabular data models are quick and easy to create compared to multidimensional data models, although
they lack some of the advanced features of multidimensional models. Tabular models are suitable for
anything from a personal desktop business intelligence (BI) application that has been developed in Excel,
to departmental or larger solutions, depending upon the complexity of the application.

Tabular models offer two principal advantages over multidimensional models:

 The relational model is widely understood and relatively intuitive, so the barrier to entry for relational
database developers and information workers wanting to develop analysis solutions, is lower for
tabular models than for multidimensional models. This helps companies to minimize costs by taking
advantage of their existing internal expertise to create data analysis solutions.

 Tabular data models are generally simpler in design than multidimensional models, so companies can
achieve a faster time to deployment for BI applications.

Creators of tabular models can use the Data Analysis Expressions (DAX) language to create measures and
calculated columns, and to implement security. DAX is similar to the formulas that are used in Excel
workbooks, so information workers who already use Excel should find it relatively easy to learn and use.

Tabular data models are generally easy to create because their structure enables developers to interact
directly with data, tables, and relationships without having to create complex additional structures, such as
dimensions and cubes. Tabular data models help to reduce development effort, enabling developers to
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 7-3

get applications into production more quickly, and making it possible for companies to take advantage of
a wider pool of talent to develop solutions in SQL Server Analysis Services. Relational models are generally
more widely understood than their multidimensional equivalents, so in-house relational database
developers should quickly adapt to creating tabular data models with minimal training.
You can create tabular data models by using the PowerPivot add-in in Excel, or by using SQL Server Data
Tools. You can use SQL Server Management Studio to manage tabular data model databases in SQL
Server Analysis Services after deploying them.

Note: This module focuses on tabular data models in databases in SQL Server Analysis
Services, which are typically created by BI developers as part of a managed enterprise BI solution.
Information workers can use the PowerPivot add-in to create their own tabular data models in
Excel workbooks for personal analysis or to share with colleagues in Microsoft SharePoint®
Server. If these PowerPivot data models become important to the business in the future, they can
be imported into a tabular database in SQL Server Analysis Services and managed centrally by IT.
For more information about PowerPivot, you can attend course number 20778B, Analyzing Data
with Power BI.

Options for Creating a Tabular Data Model Project in SQL Server Analysis
Services
A tabular data model in SQL Server Analysis
Services is a database that is stored on an instance
of SQL Server Analysis Services, running in tabular
mode. You can install an instance of SQL Server
Analysis Services in multidimensional, PowerPivot
for SharePoint, or tabular mode. To create tabular
databases, you must install at least one instance of
SQL Server Analysis Services that runs in tabular
mode.

Note: You can install multiple instances of

SQL Server Analysis Services on the same
computer so that you have different modes available.

You can create a tabular data model project for SQL Server Analysis Services by using the SQL Server Data
Tools project templates. Each of the three tabular data model templates enables you to create a project in
a different way:

 Analysis Services Tabular Project. This template creates a new, empty tabular data model. You must
import data and metadata from your data sources to populate the model.

 Import from PowerPivot. This template creates a new tabular data model by using the tabular data
model that is embedded in a PowerPivot for Excel workbook. The Import from PowerPivot option
extracts the data and metadata from the specified workbook and uses this to populate the new
model.
 Import from Server (Tabular). This template creates a new tabular data model from an existing
tabular data model in SQL Server Analysis Services. It extracts the data and metadata from the
specified database and uses these to populate the new model.
 MCT USE ONLY. STUDENT USE PROHIBITED
7-4 Implementing a Tabular Data Model by Using SQL Server Analysis Services

Before you create a tabular data model project, you must ensure that there is an instance of SQL Server
Analysis Services running in tabular mode that is available for the project to use during development. The
instance of SQL Server Analysis Services can be local or located on a network. A local instance will typically
provide better performance during the development phase, and makes it possible to work with the model
offline. However, a network instance offers easier collaboration if several developers are involved,
particularly if the project is integrated with Team Foundation Server.

You can also create a tabular data model by using the Restore from PowerPivot option in SQL Server
Management Studio. This method enables you to create a tabular data model database directly from a
PowerPivot for Excel workbook, without having to use SQL Server Data Tools. Like the Import from
PowerPivot template in SQL Server Data Tools, the Restore from PowerPivot option extracts the data
and metadata from the specified PowerPivot for Excel workbook and uses this to populate the tabular
data model database.

Key Features
The key features that you can configure in a tabular database in SQL Server Analysis Services are:
 DirectQuery mode, which passes queries directly to the underlying data source instead of using the
in-memory storage engine.

 Row-level and object-level security, which enable you to implement security at a granular level.
 Partitions, which enable you to manage large tables more efficiently.

Using SQL Server Data Tools to Develop a Tabular Data Model in SQL
Server Analysis Services
SQL Server Data Tools is based on Microsoft Visual
Studio®. It is a comprehensive single
development environment for all of your BI
projects, including projects in SQL Server
Reporting Services, and multidimensional and
tabular projects in SQL Server Analysis Services. It
supports a range of features to enable you to
manage the full project development life cycle,
including:

 Templates to create projects.

 Source control, if it is used in conjunction with

Team Foundation Server or other third-party
source control plug-ins.
 Debugging tools.

 Building and deployment directly to test and production servers.

When you create a tabular data model in SQL Server Data Tools, the tabular model designer opens
automatically. It consists of several windows, including:

 Solution Explorer. This window displays the project and its contents. By default, the contents are the
References object and the Model.bim file. You can set the properties of the project, such as the
deployment server or the query mode, from this window.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 7-5

 Designer window. This window displays a visual representation of the model, offering two different
views:

o Data view. This view displays one table at a time, showing the columns and the data that the
table contains. You can select the table that you want to view by clicking the appropriate tab. The
data view contains the measure grid, which you can use to create measures and key performance
indicators (KPIs). You can also create calculated columns in the data view by adding a DAX
expression to the formula bar.
o Diagram view. This view presents the tables in a schema-like diagram. It shows the columns and
hierarchies in each table, and the relationships between the tables. You can also use this view to
create and manage hierarchies.

 Properties window. This window displays the properties of the object that you select.

 Error list. This window displays errors and other messages that relate to the model.

 Output window. This window displays status information that relates to builds and deployments.

The tabular model designer also contains extra menu items, including Model, Table, and Column.
After creating a tabular data model in SQL Server Data Tools, you can import the necessary data and
metadata by using the Get Data functionality. This creates all of the metadata structures that underpin a
tabular data model. You do not need to create data source views, dimensions, or cubes.

The Workspace Database

Creating a tabular data model project by using
one of the templates in SQL Server Data Tools
automatically adds a new database to the tabular
instance of SQL Server Analysis Services that is
configured as the workspace server. Every tabular
data model project has its own exclusive
workspace database. You can view the workspace
database on the SQL Server Analysis Services
server, using SQL Server Management Studio.
Workspace databases use a naming convention
that combines the project name, the name of the
user who created it, and a system-generated GUID
in the following format:

<project name>_<user name>_<GUID>

This naming format ensures that every workspace database has a unique name.

The workspace database is an in-memory database that contains all of the data and metadata for the
project while it is being developed. All of the changes that you make to the model in SQL Server Data
Tools, such as creating hierarchies or adding linked tables, result in updates to the workspace database.
Deploying the model creates a new database—a renamed copy of the workspace database—on the
designated instance of SQL Server Analysis Services.

Local and Remote Workspace Databases

Ideally, the workspace database should be located on the same computer where the developer creates the
model because this is likely to result in the best performance. You can use a remote instance of SQL Server
Analysis Services to host the workspace database, but this configuration has the following limitations:

 You cannot create tabular data model projects by using the Import from PowerPivot template.
 MCT USE ONLY. STUDENT USE PROHIBITED
7-6 Implementing a Tabular Data Model by Using SQL Server Analysis Services

 You cannot use the Backup to disk option in the Data Backup property.

 You might experience slower performance because of the latency introduced by using a remote
server.

Configuring the Workspace Database

The Properties window of the Model.bim file in SQL Server Data Tools enables you to configure the
workspace database.

 The Workspace Server property enables you to specify the instance of SQL Server Analysis Services
that hosts the workspace database.

 The Workspace Retention property defines what happens to the in-memory workspace database
when you close a tabular data model project. The default Workspace Retention setting is Unload
from memory, which removes the database from memory and stores it on disk. This setting frees up
memory, but the model takes longer to load when you reopen the project. You can also configure the
Workspace Retention setting to keep the workspace database both in memory and on disk when
you close a project, or to delete it completely. If you choose this last option, the workspace database
is recreated every time you open the project.

Note: The default host for new workspace databases is defined in the Default workspace
server setting on the Workspace Database page. To access the Workspace Database page, on
the Tools menu, click Options, and then, in the Options dialog box, expand Analysis Services
Tabular Designers, and click Workspace Database.

Question: What are the advantages and disadvantages of tabular models when compared to
multidimensional models?
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 7-7

Lesson 2
Creating a Tabular Data Model
A tabular data model consists of multiple tables, usually linked by using relationships. Columns in the
tables can be used as attributes and measures that enable business users to aggregate key business
values. You can organize the attributes by which values are aggregated into hierarchies to enable drill-
down and drill-up analysis.

Lesson Objectives
After completing this lesson, you will be able to:

 Import tables into a tabular data model project.

 Define measures in a tabular data model.

 Manage relationships between tables in a tabular data model.

 Configure columns in a tabular data model.

 Create hierarchies in a tabular data model.

Importing Tables
The first step in creating a tabular data model is to
import tables of data from one or more sources.
Tabular models using compatibility level of 1400
or higher use the Get Data feature to specify the
data to use. The available data sources for a
tabular model include:
 File

o Excel

o Text/CSV

o XML

o JSON

 Database

o SQL Server database

o SQL Server Analysis Services database

o Oracle database

o MySQL database

 Azure®

o Azure SQL database

o Azure SQL data warehouse

 Online Services

o Microsoft Exchange Online

o Dynamics 365 (online)
 MCT USE ONLY. STUDENT USE PROHIBITED
7-8 Implementing a Tabular Data Model by Using SQL Server Analysis Services

 Other

o ODBC

o OLE DB

When importing from a database, you can filter tables to exclude columns that are not required for your
analysis. You should always aim to only import the necessary data into the data model to ensure optimal
performance and simplify the end-user experience. You should also provide a friendly name, or alias, for
the tables that you import. Tables in databases do not usually have easy-to-understand names, so
providing a friendly name will make the data easier for the end user to interpret.

You can import tables by using the Get Data feature in SQL Server Data Tools. To access this, on the
Model menu, click Import From Data Source.

Defining Measures
Measures are numeric values in your data model
that users can aggregate to analyze business
performance. Unlike traditional multidimensional
data models, in which measures can only be
defined as part of a measure group—usually
based on a fact table—tabular data models enable
you to define measures for any numeric column in
any table.

Measures are aggregated across multiple

dimensions within the data model, and typically
provide a summary value for core business
metrics. For example, a data model that includes a
sales order table will typically define measures that sum sales revenue, cost, and order quantities by
attributes of business entities that are defined in other tables, such as customers, products, and stores.

You can define measures in the Measure Grid area of the tabular model designer in SQL Server Data
Tools. Measures are usually based on a simple DAX function that is applied to a numeric column. For
example, a measure named Revenue could be defined as the sum of values in the SalesAmount column
by using the DAX expression Revenue:=Sum([SalesAmount]).

Sum is the most common aggregate function that is used to define measures, but you can also use
Average, Count, Max, Min, and many others. You can also define more complex measures that reference
multiple columns, or use custom algorithms.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 7-9

Demonstration: Creating a Tabular Data Model Project

In this demonstration, you will see how to:

 Create a tabular data model project.

 Import tables into a tabular model.

 Define measures in a data model.

 Test the data model in Excel.

Demonstration Steps
Create a Tabular Data Model Project

1. Ensure that the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and then
log on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd.

2. In the D:\Demofiles\Mod07 folder, run Setup.cmd as Administrator.

3. In the User Account Control dialog box, click Yes.

4. Start SQL Server Data Tools 2015, and then on the File menu, point to New, and then click Project.
5. In the New Project dialog box, type the following values, and then click OK:

o Project Template: Analysis Services Tabular Project

o Name: TabularDemo
o Location: D:\Demofiles\Mod07

6. If the Tabular model designer dialog box is displayed, type the following values, and then click OK:

o Workspace server: localhost\SQL2

o Compatibility level: SQL Server 2017 / Azure Analysis Services (1400)

7. On the Tools menu, click Options.

8. In the Options dialog box, expand Analysis Services Tabular Designers, and then click Workspace
Database. Note the name of the default workspace server that is used to host the data model during
development, and then click Cancel.

Import Tables into a Tabular Model

1. On the Model menu, click Import From Data Source.

2. In the Get Data window, select SQL Server database, and then click Connect.

3. On the SQL Server database page, type the following values, and then click OK:

o Server: MIA-SQL

o Database (optional): AdventureWorksDW

4. On the Impersonation Information page, change the Impersonation Mode to Impersonate Service
Account, then click Connect.

5. In the Encryption Support dialog box, click OK

6. On the Navigator page, select the following source tables, and then click OK:

o DimCustomer

o DimDate

o DimGeography
 MCT USE ONLY. STUDENT USE PROHIBITED
7-10 Implementing a Tabular Data Model by Using SQL Server Analysis Services

o DimProduct

o DimProductCategory

o DimProductSubcategory

o FactInternetSales
7. In the Query Editor, in the Queries list, click DimCustomer table and change the Name property to
Customer.

8. On the Columns menu, click Choose Columns, clear the check box for each of the following
columns, and then click OK:

o Title

o MiddleName

o NameStyle

o Suffix

o SpanishEducation

o FrenchEducation

o SpanishOccupation

o FrenchOccupation

9. In the Queries list, click DimDate and change the Name property to Date.
10. On the Columns menu, click Choose Columns, clear the check box for each of the following
columns, and then click OK:

o SpanishDayNameOfWeek

o FrenchDayNameOfWeek

o SpanishMonthName

o FrenchMonthName

o CalendarSemester

o FiscalSemester

11. In the Queries list, click DimGeography and change the Name property to Geography.

12. In the Queries list, click DimProduct and change the Name property to Product.

13. In the Queries list, click DimProductCategory and change the Name property to Product
Category.

14. In the Queries list, click DimProductSubCategory and change the Name property to Product
Subcategory.

15. In the Queries list, click FactInternetSales and change the Name property to Internet Sales.

16. On the toolbar, click Import, and then wait for the data to be imported. When the data has been
imported successfully, click Close.

17. Click each of the tabs in the model designer to see the data that has been imported into each table.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 7-11

Define Measures in a Data Model

1. Click the Internet Sales tab, and then click the first empty cell in the measure grid under the
SalesAmount column.

2. On the Column menu, point to AutoSum, and then click Sum.

3. In the formula bar, modify the expression that has been generated to change the name of the
measure to Revenue, as shown in the following example:

Revenue:=SUM([SalesAmount])

4. Widen the SalesAmount column, and then note that the measure is calculated and displayed as a
currency value. The formatting has been inherited from the column.

5. Select the cell that contains the Revenue measure, and then press F4 to display the properties pane.
View the Format property, which is set to Currency.

6. Create a second measure named Cost in the measure grid under the TotalProductCost column
based on the following expression, and format the Cost measure as currency:

Cost:=Sum([TotalProductCost])

7. On the File menu, click Save All.

Test the Data Model in Excel

1. On the Model menu, click Analyze in Excel.

2. In the Analyze in Excel dialog box, ensure that Current Windows User is selected and that the
(Default) perspective is specified, and then click OK.

3. When Excel opens, in the PivotTable Fields pane, note that the measures that you defined are
displayed under an Internet Sales measure group. The Internet Sales table is also shown in this
pane, and contains the columns in the table, including those on which the aggregated measures are
based. Some users might find this confusing so, in a future refinement of the data model, the columns
that are not useful for analysis should be hidden.

4. Select the Revenue measure so that it is summarized in the PivotTable.

5. In the Date table, select CalendarYear so that the revenue in the PivotTable is aggregated by year.
Note that there is no indication in the Date table that tells the user if the values that are displayed
represent the order date, the due date, or the ship date that is associated with the order.
6. In the Date table, select EnglishMonthName so that the revenue is further broken down into
monthly totals. Note that the months are shown in alphabetical order instead of chronological order.

7. Close Excel without saving the workbook.

8. Keep SQL Server Data Tools open for the next demonstration.
 MCT USE ONLY. STUDENT USE PROHIBITED
7-12 Implementing a Tabular Data Model by Using SQL Server Analysis Services

Managing Relationships
When you import data from a table in a relational
database, you can choose to automatically import
additional tables that have a relationship with that
table. The import process identifies related tables
by analyzing their foreign-key relationships.
However, data analysts frequently need to analyze
data that comes from multiple sources and for
which explicit relationships are not defined. You
can manually create, edit, and delete relationships
as required in tabular data model projects.

When you create a relationship, you must specify

the two tables that you want to relate and the
common column that contains the key values. Relationships between tables in tabular data models are
one-to-many, so, when you create a relationship, you must specify the columns and tables for each end of
it. To create a relationship, in the diagram view of the data model designer, drag the column in one of the
tables in the relationship to the column in the other table in the relationship.

Tabular data models support multiple relationships between tables. For example, date and fact tables in
data warehouses often have multiple relationships such as order date, due date, and delivery date.
However, only one relationship is marked as being active between the tables. When you view multiple
relationships in the diagram view of the data model designer, the active relationship shows as an
unbroken line, and inactive relationships appear as dotted lines.
If you need to implement a “role-playing dimension”, in which multiple relationships between a fact table
and a dimension table must be active concurrently, you must import the dimension table multiple times,
assigning a different name each time, and create individual relationships between the fact table and each
copy of the dimension table. For example, if you needed to implement order date, due date, and delivery
date role-playing dimensions based on a date dimension table, you would need to import the date
dimension table three times, assigning an appropriate name to each copy of the table. You would then
create three individual relationships between each of the keys that were used for order date, due date,
and delivery date in the fact table, and the appropriate copy of the date dimension table.

Configuring Columns
The columns in the tables that you import into a
tabular data model become the attributes by
which analysts can aggregate measures for
analysis. For example, a Customer table might
include City and Gender columns, enabling
analysts to aggregate numerical measures such as
revenue by city, or order quantity by customer
gender.

Specifying Column Data Types

When you have imported the tables for your data
model, you can specify the data type of each
column in the Properties window. For example, if
the Customer table includes a DateOfBirth column, you can specify that it contains date values.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 7-13

Creating Calculated Columns

In addition to columns in the source table, you can define additional attributes by creating calculated
columns. A calculated column is based on a DAX formula, which uses a similar syntax to a formula in Excel.
For example, in the Customer table that was described previously, you could concatenate the FirstName
and LastName columns to create a calculated column named FullName, based on the DAX expression
=[FirstName] & " " & [LastName], or create a calculated column named YearOfBirth, based on the DAX
expression =YEAR([DateOfBirth]).

Specifying Sort Order for Columns

You can define sort orders for columns, including sorting a column based on the value in another column
in the same table. This approach is most commonly used in time dimension tables that contain numeric
values—such as a MonthOfYear column that has values between 1 and 12—and text values, such as a
MonthName column that has values from “January” to “December.” In this example, you can specify that
the MonthName column should be sorted based on the corresponding MonthOfYear value. You can
define a sort order for a column in the Sort by Column dialog box, by clicking the column heading in the
data view of the data model designer, and then, on the Column menu, pointing to Sort, and clicking Sort
by Column.

Hiding and Renaming Columns

By default, all columns in all tables are visible when the data model is browsed in a client tool such as
Excel. This includes columns that are not useful for aggregation, such as those on which you have based
measures, and columns that are encapsulated in hierarches. You can make the data model simpler for
users to browse by hiding columns that are not useful as attributes for aggregation. You can also rename
the columns that you want to show in the client tool to a more user-friendly name. To hide or rename a
column in the diagram view of the data model designer, maximize the table that contains the column,
right-click the column, and then click Hide from Client Tools or Rename.

Demonstration: Managing Relationships and Columns

In this demonstration, you will see how to:

 Manage relationships.

 Rename and hide columns.

 Configure column sort order.

 Review your changes in Excel.

Demonstration Steps
Manage Relationships

1. Ensure that you have completed the previous demonstration in this module.

2. In SQL Server Data Tools, with the Model.bim pane visible, on the Model menu, point to Model
View, and then click Diagram View. The tables are shown as a schema diagram, with lines between
them to denote relationships. Note that there are three relationships between the Internet Sales
table and the Date table.

3. Double-click the solid line between the Internet Sales and Date tables. Note the columns that are
used to define this relationship, and that this is an active relationship, and then click Cancel.

4. Double-click each of the dotted lines between the Internet Sales and Date tables, and then click
Cancel. In each case, note the columns that are used to define these relationships and that these are
inactive relationships, and then click Cancel.
 MCT USE ONLY. STUDENT USE PROHIBITED
7-14 Implementing a Tabular Data Model by Using SQL Server Analysis Services

5. Right-click each dotted relationship line in turn, and then click Delete. When you are prompted to
permanently delete the relationship from the model, click Delete from Model.

6. Double-click the remaining solid line between the Internet Sales and Date tables.

7. In the Edit Relationship dialog box, in the Columns list for the Internet Sales table, select
OrderDateKey and then click OK (if it is already selected, click Cancel), This ensures that the active
relationship is based on the order date, not the shipping date or due date.

8. Right-click the Date table title bar, click Rename, and then rename the table to Order Date.

9. In the Tabular Model Explorer, expand Data Sources, right-click SQL\mia-sql:AdventureWorksDW,

and then click Import New Tables.

10. On the Navigator page, select the DimDate table, and then click OK.

11. In the Query Editor, in the Queries list, click DimDate table and change the Name to Ship Date.

12. On the Columns menu, click Choose Columns, clear the following columns, and then click OK:

o SpanishDayNameOfWeek

o FrenchDayNameOfWeek
o SpanishMonthName

o FrenchMonthName

o CalendarSemester
o FiscalSemester

13. Click Import, and then wait for the table to be imported. When it has been imported successfully,
click Close.

14. Arrange the diagram so that you can see the Ship Date and Internet Sales tables, and then drag the
ShipDateKey column from the Internet Sales table to the DateKey column in the Ship Date table
to create the relationship. If the Ship Date table does not appear, refresh the page.
Rename and Hide Columns

1. Click the title bar of the Order Date table, and then click its Maximize icon.
2. In the maximized Order Date table, click the DateKey column, hold down CTRL, click the
DayNumberOfWeek and MonthNumberOfYear columns to select them, right-click any of the
selected columns, and then click Hide from Client Tools.

3. In the maximized Order Date table, right-click the FullDateAlternateKey column, click Rename, and
then rename the column to Date.

4. Repeat the previous step to rename the following columns:

o EnglishDayNameOfWeek (rename to Weekday)

o EnglishMonthName (rename to Month)

5. Click the Restore icon for the Order Date table.

6. Maximize the Internet Sales table, hide all columns other than the Revenue and Cost measures that
you created in the previous exercise, and then restore the Internet Sales table.

Configure Column Sort Order

1. On the Model menu, point to Model View, and then click Data View.

2. On the Order Date tab, click the Weekday column heading.

 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 7-15

3. On the Column menu, point to Sort, and then click Sort by Column.

4. In the Sort by Column dialog box, in the Sort column, ensure that Weekday is selected, in the By
column, select DayNumberOfWeek, and then click OK.

5. Click the Month column heading.

6. On the Column menu, point to Sort, and then click Sort by Column.
7. In the Sort by Column dialog box, in the Sort column, ensure that Month is selected, in the By
column, select MonthNumberOfYear, and then click OK.

8. On the File menu, click Save All.

Review Your Changes in Excel

1. On the Model menu, click Analyze in Excel.

2. In the Analyze in Excel dialog box, ensure that Current Windows User is selected and that the
(Default) perspective is specified, and then click OK.

3. When Excel opens, in the PivotTable Fields pane, note that the measures that you defined are still
displayed under an Internet Sales measure group, but that the Internet Sales table is no longer
shown here because all other columns have been hidden.

4. Select the Revenue measure so that it is summarized in the PivotTable.

5. In the Order Date table, select CalendarYear so that the revenue in the PivotTable is aggregated by
year based on the order date.

6. In the Order Date table, select Month so that the revenue is further broken down into monthly
totals. Note that the months are now shown in chronological order.
7. In the Order Date table, select CalendarQuarter, and then note that the quarter is shown under
each month.

8. In the PivotTable Fields pane, in the Rows area, drag CalendarQuarter so that it is above Month,
and the PivotTable shows the correct hierarchy for year, quarter, and month. This user experience
could be improved by enabling users to select a hierarchy that automatically enables them to drill
down to the correct levels.
9. Close Excel without saving the workbook.

10. Keep SQL Server Data Tools open for the next demonstration.

Creating Hierarchies
A hierarchy is a collection of attributes from a
single table and it is organized as parent and child
nodes. You create hierarchies to simplify the end-
user experience and enable drill-up and drill-down
aggregations. For example, you might have
Product Category, Subcategory, and Product
Name columns in a table. These values represent a
natural hierarchy in which you can organize data,
enabling users to view aggregations such as total
sales revenue at the Product Category, Product
Subcategory, and Product Name levels. Another
 MCT USE ONLY. STUDENT USE PROHIBITED
7-16 Implementing a Tabular Data Model by Using SQL Server Analysis Services

common hierarchy is Year, Quarter, and Month in a time dimension table.

You can manage hierarchies for data in a table by maximizing the table in the diagram view of the data
model designer. To create a hierarchy, on the title bar of the maximized table, click the Create Hierarchy
icon. To add a column to a hierarchy, in the maximized table, right-click the column, point to Add to
Hierarchy, and then click the name of the hierarchy.

Hierarchies with a differing number of levels are called ragged hierarchies. By default, these hierarchies
display blanks for levels with no data. You can set the Hide Members property of the hierarchy to Hide
blank members to stop displaying these levels.

Demonstration: Creating a Hierarchy

In this demonstration, you will see how to:

 Create a hierarchy.

 Browse the hierarchy in Excel.

Demonstration Steps
Create a Hierarchy

1. Ensure that you have completed the previous demonstrations in this module.
2. In SQL Server Data Tools, with the Model.bim pane visible, on the Model menu, point to Model
View, and then click Diagram View.

3. Maximize the Order Date table, and then on its title bar, click the Create Hierarchy icon. Name the
new hierarchy Calendar Date.
4. Drag the CalendarYear column to the Calendar Date hierarchy.

5. Drag the CalendarQuarter column to the Calendar Date hierarchy.

6. Drag the Month column to the Calendar Date hierarchy.

7. Drag the Date column to the Calendar Date hierarchy.

8. Restore the Order Date table, and then on the File menu, click Save All.

Browse the Hierarchy in Excel

1. On the Model menu, click Analyze in Excel.

2. In the Analyze in Excel dialog box, ensure that Current Windows User is selected and that the
(Default) perspective is specified, and then click OK.

3. When Excel opens, select the Revenue measure so that it is summarized in the PivotTable.

4. In the Order Date table, select Calendar Date so that the revenue in the PivotTable is aggregated by
year.

5. In the PivotTable, expand the years to show the revenue by quarter, expand the quarters to view
revenue by month, and then expand the months to view revenue for individual dates.

6. Close Excel without saving the workbook.

7. Keep SQL Server Data Tools open for the next demonstration.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 7-17

Verify the correctness of the statement by placing a mark in the column to the right.

Statement Answer

You are implementing an order

date and a ship date role-playing
dimension based on a date
dimension table. You can achieve
this by importing the fact table
and the date dimension table
once, and then creating a
relationship between the order
date key in the fact table and the
date key in the date dimension
table, and also creating a
relationship between the ship
date key in the fact table and the
date key in the date dimension
table.
 MCT USE ONLY. STUDENT USE PROHIBITED
7-18 Implementing a Tabular Data Model by Using SQL Server Analysis Services

Lesson 3
Using a SQL Server Analysis Services Tabular Model in an
Enterprise BI Solution
SQL Server Analysis Services includes tabular data model capabilities that can be useful in an enterprise BI
scenario. It enables you to define perspectives as subsets of the data model for specific user groups, and
partition the tables in the data model to improve data refresh processing performance. There is also a
choice of query modes to help you to optimize the right balance of query and processing performance,
and the ability to define security restrictions on a data model.

Lesson Objectives
After completing this lesson, you will be able to:

 Create perspectives in a tabular data model.

 Create partitions in a tabular data model.

 Process tabular databases, tables, and partitions.

 Implement DirectQuery mode in a tabular data model.

 Implement security in a tabular data model.

 Deploy a tabular data model.

Perspectives
Tabular data models can contain large numbers of
tables and measures, which can make it difficult
for users to work with the data in a client tool such
as Excel. You can simplify the user experience by
creating perspectives that display only a subset of
the tables, columns, and measures in a model. For
example, the tables in your tabular data model
might contain many different types of data, such
as human resources, sales, products, promotions,
and financial data. However, users who want to
analyze sales performance will probably not need
to view the financial or human resources tables, so
you could create a perspective where these are explicitly excluded. When users connect to the model with
this perspective, they will not see the excluded tables in their client application.

Creating Perspectives
You can create a perspective in a tabular data model by using the Perspectives dialog box in SQL Server
Data Tools. To open this dialog box, on the Model menu, click Perspectives, and then click Create and
Manage.

When you create a perspective, you must provide a name for it and specify the tables, measures, and
columns to include. When users connect to the model, they will be able to select the perspective that they
want to use. You should ensure that the names that you give to perspectives are descriptive and easy to
understand. To speed up the process, you can also create a perspective by copying an existing similar one.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 7-19

Perspectives and Security

You cannot use perspectives to secure tabular data models. You cannot define permissions on a
perspective, so users who have the right to view the tabular data model will be able to use any
perspective that is defined in the model. In addition, if a user does not have permissions to view certain
tables, they cannot circumvent this by using a perspective. If you need to secure your tabular data models,
you should create roles and define permissions for the required objects. You will learn about security for
tabular data models later in this module.

Partitions
Partitions enable you to manage large tables more
efficiently. You can divide a table into logical units
that you can load or reload individually. For
example, you could use a large table called Sales
in your tabular data model with a data warehouse
fact table, which is updated daily, as the source. In
this example, the data would quickly go out of
date. You could refresh the data in Sales, but this
operation is time-consuming because it reloads all
of the rows from the source table. To keep Sales
synchronized with the source table more
efficiently, you could create partitions that divide
up the table, perhaps by using the month or year of the order date. Then, you would only need to
regularly refresh the partition that contains the most recent data, updating the other partitions only when
required.

Note: Partitions in tabular data models are not intended to improve query performance in
the way that partitions in other types of databases can. Partitioning can, however, improve the
performance of data model processing by enabling you to process only partitions that contain
new or updated data.

Creating Partitions
Every table that you import into a tabular data model has a single partition that is associated with it by
default. This enables you to refresh each table separately. During the development stage, you can create a
new partition in a tabular data model by performing the following steps:
1. In SQL Server Data Tools, in the tabular model designer, on the Table menu, and then click
Partitions.

2. In the Partition Manager dialog box, select the table for which you want to create a partition, and
then click New. You can also create a partition by copying an existing one.

3. To select the rows to include in the partition, use the Table Preview pane, or use the Query Editor
pane to define a Transact-SQL query. The partitions that you create in this way are added to the
workspace database for the project.

For tabular data models that are deployed to an instance of SQL Server Analysis Services, you can create
and manage partitions by using SQL Server Management Studio.
 MCT USE ONLY. STUDENT USE PROHIBITED
7-20 Implementing a Tabular Data Model by Using SQL Server Analysis Services

Using Partitions in the Development Phase

When you are creating a tabular data model by using a source database that has large tables, you should
consider implementing partitions to enable you to work with a reduced sample data set. Using a sample
data set during development is more efficient because it reduces data load times and enables better
query performance. You can create, delete, edit, and merge partitions in the workspace database as
required. When you deploy the model, you can import all of the data and employ a different partitioning
design that is appropriate to the deployed database.

Analysis Services Processing Types

Processing populates the tables and partitions in a
tabular data model by importing data from the
source database. You can choose to process
individual tables or partitions, or all of the tables
and partitions in a model together. Depending on
the option that you choose, processing rebuilds
relationships and hierarchies in addition to
recalculating calculated columns and measures.

The following processing types are available:

 Process Default. This option detects the

current state of a table or partition and
performs the necessary actions to ensure that
it is in a fully processed state. When you first create a table or partition, this option loads it and
creates all hierarchies, relationships, measures, and calculated columns.
 Process Full. This option fully processes a table or partition. It removes existing data and repopulates
partitions, before recreating hierarchies, relationships, measures, and calculated columns.
 Process Data. This option only processes data—it does not recreate hierarchies, relationships,
measures, and calculated columns.

 Process Clear. This option deletes data from a table or partition.

 Process Add. This option updates tables or partitions by only adding new data.

You can process partitions by using the Process Partitions dialog box. To access the Process Partitions
dialog box in SQL Server Data Tools, on the Model menu, point to Process, and then click Process
Partitions. To access the Process Partition(s) dialog box in SQL Server Management Studio, right-click
the table that you need to update, point to Partitions, and then click Process.

You can also use SQL Server Management Studio to process tabular databases, tables, or partitions. For
example, to process a database, right-click the database, and then click Process Database, select a
processing type, and then click OK.

Note: When you process a table or database with more than one partition, the partitions
are processed in parallel in order to maximize performance. By using the TMSL Sequence
command, you can specify sequential processing, in which each partition is processed in turn but
this is beyond the scope of this course.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 7-21

For more information about the Sequence command, see Sequence command (TMSL) in Microsoft Docs:

Sequence command (TMSL)

https://aka.ms/ogen8x

DirectQuery Mode
The xVelocity engine provides in-memory storage
and data processing for very fast query handling,
and is appropriate for most tabular data models in
SQL Server Analysis Services. However, you can
also configure a tabular data model to bypass the
xVelocity engine and route queries directly to the
underlying data source by using DirectQuery
mode. This option might be suitable in the
following circumstances:

 For tabular data models that have very large

data sets that take a long time to load into
memory, and which might fail to load if there
is insufficient memory available. Using DirectQuery removes the need to load data into memory.

 For tabular data models that need to contain up-to-date information and for which frequent
processing is impractical. DirectQuery retrieves source data directly, providing access to the most
recent data.
 For organizations that have invested in high-specification data warehouse hardware that can deliver
excellent performance. DirectQuery enables companies to take advantage of their hardware
performance while still providing access to data through a tabular data model. In this situation, you
should test thoroughly to determine whether you can achieve the best performance by relying on the
data warehouse hardware or by using the xVelocity engine.

Designing Tabular Data Models for Use with DirectQuery

When you design tabular data models for which you intend to enable DirectQuery mode, there are
several points to consider:

 DirectQuery only supports a single data source, which must be a SQL Server relational database, an
Oracle relational database, or a Teradata relational database.
 You cannot use certain DAX functions, such as those for time intelligence, in the model. Others, such
as certain statistical functions, might behave differently in DirectQuery mode.
 There are limitations to the Multidimensional Expressions (MDX) that you can use in DirectQuery
mode.

For further details about the limitations of features that you can use in DirectQuery mode, see the
Restrictions section of DirectQuery Mode (SSAS Tabular) in Microsoft Docs:
DirectQuery Mode (SSAS Tabular)
http://aka.ms/M6m6uz
 MCT USE ONLY. STUDENT USE PROHIBITED
7-22 Implementing a Tabular Data Model by Using SQL Server Analysis Services

Enabling DirectQuery Mode

To enable DirectQuery for a database in SQL Server Management Studio, configure the Default Mode
setting on the Model page of database properties. To enable DirectQuery in SQL Server Data Tools,
configure the DirectQuery Mode setting in the properties of the Model.bim file.

Security for Tabular Data Models

You can configure security in a tabular data model
by creating database roles, and then applying
permissions and filters to them. You should use
roles to gather together Windows® user accounts
and groups that have the same data access
requirements, and then apply the permissions to
the role. You can create and manage roles in a
tabular data model in the development phase by
using the Role Manager dialog box in SQL Server
Data Tools. To access this dialog box, on the
Model menu, click Roles.

Database Permissions
You can use Role Manager to define database permissions for roles. You can define permissions that allow
five levels of access to a tabular data model database. They are:

 Read. This permission gives role members read access to the data, but they cannot see the model in
SQL Server Management Studio.

 Read and Process. This permission gives role members read access to the database and allows them
to process data, but they cannot see the model in SQL Server Management Studio.

 Process. This permission gives role members the ability to process data in the database.

 Administrator. This permission gives role members rights to read, process, and modify the data.
 None. This permission prevents role members from accessing the data.

You can also define database permissions for roles by using SQL Server Management Studio.

Row-Level Security
You can create DAX expressions that filter the rows that role members can see. DAX filters are expressions
that evaluate to TRUE or FALSE. Rows that evaluate to TRUE are visible to role members. For example, the
expression in the following code example filters the Region column in the Sales Territory table:

='Sales Territory'[Region]="France"

The rows that contain the value France evaluate to TRUE, so they are visible to role members. Rows that
contain other values in the Region column are not visible.
You can only define DAX filters for roles that have the Read and Read and Process database permissions.
Roles that have None or Process permissions cannot view any data, so filters are irrelevant. Roles that
have the Administrator permission can view all data. You can test the database permissions and row-
level filters that you define by using the Analyze in Excel option in SQL Server Data Tools, and
connecting as a specific role.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 7-23

Object-Level Security
You can also use object-level security for tables and columns to restrict access to sensitive data and object
names when you do not want a user to know that a table exists. You configure object-level security using
JSON-based metadata, Tabular Model Scripting Language (TMSL), or Tabular Object Model (TOM), setting
the metadataPermission property of the tablePermissions or columnPermissions object.

Deploying a Tabular Data Model

When you have completed the development
phase for a tabular data model, you can deploy it
to an instance of SQL Server Analysis Services that
is running in tabular mode. This enables
information workers and data analysts to connect
to the data model by using client tools such as
Excel and Power View. Deploying a tabular data
model copies the data and metadata in the
workspace database and uses it to create a new
database that has the same name as the project by
default. You can deploy to the same server that
you used to develop the model, but, more often,
you will deploy to a different instance of SQL Server Analysis Services, such as a test or production server.

You can control this process by using the Deployment Options and Deployment Server settings in the
project properties in SQL Server Data Tools. To access the project properties, in Solution Explorer, right-
click the project, and then click Properties.

 There are two options available in the Deployment Options settings:

 Processing Option. Use this setting to define how the new database is processed. There are three
options:

o Default. When you select this option, SQL Server Analysis Services detects the state of the
database and processes the data and metadata accordingly.

o Full. When you select this option, SQL Server Analysis Services processes all data and metadata.

o Do Not Process. When you select this option, SQL Server Analysis Services deploys only the
metadata.

 Transactional Deployment. By default, SQL Server Analysis Services will deploy objects even if
processing for those objects fails. You can use this option to force deployment to fail if processing
fails.

 There are four options available in the Deployment Server settings:

 Server. Use this option to specify the name of the deployment server.

 Edition. Use this option to specify the edition of the instance of SQL Server Analysis Services on the
deployment server.
 Database. Use this option to specify the name of the database on the deployment server.

 Model Name. Use this option to specify the name that client tools see when they connect to the
model.
 MCT USE ONLY. STUDENT USE PROHIBITED
7-24 Implementing a Tabular Data Model by Using SQL Server Analysis Services

BI Semantic Model Connection

Clients can connect directly to a tabular data model in SQL Server Analysis Services by using tools such as
Excel. However, you can make the connection process easier for users by creating a BI semantic model
connection in the PowerPivot Gallery. A BI semantic model connection enables users to connect to a
tabular database in SQL Server Analysis Services or a PowerPivot workbook by using Excel or Power View
with a single click—and without needing to create connections manually.

For further details about how to create a BI semantic model connection, see Create a BI Semantic Model
Connection to a Tabular Model Database in Microsoft Docs:
Create a BI Semantic Model Connection to a Tabular Model Database
http://aka.ms/jc2xy0

Demonstration: Deploying a Tabular Data Model

In this demonstration, you will see how to:
 Deploy a tabular data model project.

 Use a tabular SQL Server Analysis Services database in Excel.

Demonstration Steps
Deploy a Tabular Data Model Project

1. Ensure that you have completed the previous demonstrations in this module.
2. In SQL Server Data Tools, in Solution Explorer, right-click the TabularDemo project, and then click
Properties.

3. In the TabularDemo Property Pages dialog box, type the following values, and then click OK:

o Server: localhost\SQL2
o Database: DemoDB

o Model Name: Internet Sales

4. On the Build menu, click Deploy TabularDemo. If you are prompted for impersonation credentials,
type the user name ADVENTUREWORKS\ServiceAcct and the password Pa55w.rd, and then click
OK.

5. In the Deploy dialog box, when deployment has completed, click Close.

6. Close SQL Server Data Tools.

Use a Tabular SQL Server Analysis Services Database in Excel

1. Start Excel, and then create a new blank workbook.

2. On the Data tab, in Get External Data area, in the From Other Sources drop-down list, select From
Analysis Services.

3. In the Data Connection Wizard dialog box, in the Server name box, type MIA-SQL\SQL2, ensure
that Use Windows Authentication is selected, and then click Next.

4. On the Select Database and Table page, ensure that the DemoDB database and the Internet Sales
cube are selected, and then click Next.

5. On the Save Data Connection File and Finish page, click Finish.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 7-25

6. In the Import Data dialog box, ensure that the Existing worksheet option is selected, and then click
OK.

7. In the PivotTable Fields pane, under Internet Sales, select Revenue.

8. In the PivotTable Fields pane, under Order Date, select the Calendar Date hierarchy.
9. In the PivotTable Fields pane, under Geography, drag EnglishCountryRegionName to the
Columns area.

10. Explore the data in the PivotTable. When you have finished, close Excel without saving the workbook.

Check Your Knowledge

Question

Which of the following statements

regarding DirectQuery mode is correct?

Select the correct answer.

A tabular data model that uses

DirectQuery mode will always have a
better performance than a tabular
data model that uses in-memory
storage.

A tabular data model that uses in-

memory storage will always have a
better performance than a tabular
data model that uses DirectQuery
mode.

A tabular data model that uses

DirectQuery mode will always contain
up-to-date information.

A tabular data model that uses

DirectQuery mode will only support a
SQL Server relational database data
source.

You cannot use MDX in a tabular data

model that uses DirectQuery mode.
 MCT USE ONLY. STUDENT USE PROHIBITED
7-26 Implementing a Tabular Data Model by Using SQL Server Analysis Services

Lab: Implementing a Tabular Data Model in SQL Server

Analysis Services
Scenario
Adventure Works needs an enterprise data model for reseller sales analysis, and you have decided to
implement it by using a tabular model.

Objectives
After completing this lab, you will be able to:

 Create a tabular data model project in SQL Server Analysis Services.

 Configure columns and relationships.

 Deploy a tabular data model project in SQL Server Analysis Services.

Estimated Time: 60 minutes

Virtual machine: 20768C-MIA-SQL

User name: ADVENTUREWORKS\Student

Password: Pa55w.rd

Exercise 1: Creating a Tabular Data Model Project in SQL Server Analysis

Services
Scenario
Adventure Works Cycles has a data warehouse that contains fact and dimension tables for sales. You plan
to use the tables that relate to reseller sales as the basis of your data model.

The main tasks for this exercise are as follows:

1. Prepare the Lab Environment

2. Create a Tabular Project in SQL Server Analysis Services

3. Import Tables into the Data Model

4. Create Measures

5. Test the Model

 Task 1: Prepare the Lab Environment

1. Ensure that the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and then
log on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd.

2. Run Setup.cmd in the D:\Labfiles\Lab07\Starter folder as Administrator.

 Task 2: Create a Tabular Project in SQL Server Analysis Services

1. Use SQL Server Data Tools to create a tabular project in SQL Server Analysis Services named
AWSalesTab in the D:\Labfiles\Lab07\Starter folder.
2. Use the localhost\SQL2 instance of SQL Server Analysis Services as the workspace server, and set the
compatibility level of the project to SQL Server 2017 / Azure Analysis Services (1400).
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 7-27

 Task 3: Import Tables into the Data Model

1. Open the Model.bim model, and then, using the Impersonate service account, import the
following tables from the AdventureWorksDW database on the MIA-SQL instance of SQL Server
(use the friendly names in parentheses):

o DimDate

o DimEmployee

o DimGeography

o DimProduct

o DimProductCategory

o DimProductSubcategory

o DimReseller

o FactResellerSales
2. Filter the following tables to remove the columns listed below:

 DimDate

o SpanishDayNameOfWeek

o FrenchDayNameOfWeek

o DayNumberOfYear

o WeekNumberOfYear
o SpanishMonthName

o FrenchMonthName

o CalendarSemester
o FiscalSemester

 DimEmployee

o SalesTerritoryKey
o NameStyle

o Title

o HireDate
o BirthDate

o LoginID

o EmailAddress

o Phone

o MaritalStatus

o EmergencyContactName

o EmergencyContactPhone

o SalariedFlag

o Gender

o PayFrequency
 MCT USE ONLY. STUDENT USE PROHIBITED
7-28 Implementing a Tabular Data Model by Using SQL Server Analysis Services

o Baserate

o VacationHours

o SickLeaveHours

o CurrentFlag

o SalesPersonFlag

o StartDate

o EndDate

o Status

 DimGeography

o SpanishCountryRegionName

o FrenchCountryRegionName

o IpAddressLocator

 DimProduct

o WeightUnitMeasureCode
o SizeUnitMeasureCode

o SpanishProductName

o FrenchProductName
o FinishedGoodsFlag

o SafetyStockLevel

o ReorderPoint
o DaysToManufacture

o ProductLine

o DealerPrice
o Class

o Style

o ModelName
o FrenchDescription

o ChineseDescription

o ArabicDescription

o HebrewDescription

o ThaiDescription

o GermanDescription
o JapaneseDescription

o TurkishDescription

o StartDate

o EndDate
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 7-29

o Status

 DimProductCategory

o SpanishProductCategoryName

o FrenchProductCategoryName

 DimProductSubcategory

o SpanishProductSubcategoryName

o FrenchProductSubcategoryName

 DimReseller

o OrderFrequency

o OrderMonth

o FirstOrderYear

o LastOrderYear

o ProductLine

o AddressLine1
o AddressLine2

o AnnualSales

o BankName
o MinPaymentType

o MinPaymentAmount

o AnnualRevenue
o YearOpened

 FactResellerSales

o DueDateKey
o PromotionKey

o CurrencyKey

o SalesTerritoryKey
o RevisionNumber

o CarrierTrackingNumber

o CustomerPONumber

o DueDate

 Task 4: Create Measures

1. In the model that has been created, edit the Reseller Sales table to add the following measures:

o Quantity:=Sum([OrderQuantity])

o Cost:=Sum([TotalProductCost])

o Revenue:=Sum([SalesAmount])
 MCT USE ONLY. STUDENT USE PROHIBITED
7-30 Implementing a Tabular Data Model by Using SQL Server Analysis Services

2. Format the Quantity measure as a whole number that includes a thousand separator, and ensure that
the Cost and Revenue measures are formatted as currency values.

 Task 5: Test the Model

1. Analyze the model that you have created in Excel.

2. Note the following problems with the model:

o Aggregated measures from the Reseller Sales table are easily confused with the columns on
which they are based.

o Column names are not user-friendly.

o Months in the Date table are sorted alphabetically, not chronologically.

o The relationship between Reseller Sales and Date is ambiguous, because sales orders include an
order date and a ship date.

3. You will fix these problems in the next exercise.

Results: After this exercise, you should have created a tabular data model project.

Exercise 2: Configuring Columns and Relationships

Scenario
You have created an initial tabular data model, but have identified some problems. The column names are
not user-friendly, some of the relationships are ambiguous, and some date attributes are being sorted
alphabetically rather than chronologically. You are now going to address these issues.

The main tasks for this exercise are as follows:

1. Configure Relationships
2. Rename and Hide Columns

3. Configure Column Sort Order

4. Create Hierarchies

5. Test the Model

 Task 1: Configure Relationships

1. View the existing relationships between the Reseller Sales and Date tables.

2. Make the active relationship the one based on the order date, delete all other relationships between
these tables, and then rename the Date table to Order Date.

3. Open the existing connection to the AdventureWorksDW database, and then import the DimDate
table again:

o Use the ADVENTUREWORKS\ServiceAcct account with the password Pa55w.rd if you are
prompted for impersonation credentials.

o Use the friendly name Ship Date for the DimDate table.
o Filter the table to delete the following columns:
 SpanishDayNameOfWeek
 FrenchDayNameOfWeek
 DayNumberOfYear
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 7-31

 WeekNumberOfYear
 SpanishMonthName
 FrenchMonthName
 CalendarSemester
 FiscalSemester
4. Create a relationship between the Reseller Sales and Ship Date tables based on the ShipDateKey
and DateKey columns.

 Task 2: Rename and Hide Columns

1. In the Geography table, hide the GeographyKey and SalesTerritoryKey columns, and then rename
the columns as shown in the following table:

Column New name

StateProvinceCode State or Province Code

StateProvinceName State or Province

CountryRegionCode Country or Region Code

EnglishCountryRegionName Country or Region

PostalCode Postal Code

2. In the Reseller table, hide the ResellerKey, GeographyKey, and ResellerAlternateKey columns, and
then rename the columns as shown in the following table:

Column New name

BusinessType Business Type

ResellerName Reseller Name

NumberEmployees Employees

3. In the Employee table, hide the EmployeeKey, ParentEmployeeKey,

EmployeeNationalIDAlternateKey, and ParentEmployeeNationalIDAlternateKey columns, and
then rename the columns as shown in the following table:

Column New name

FirstName First Name

LastName Last Name

MiddleName Middle Name

DepartmentName Department
 MCT USE ONLY. STUDENT USE PROHIBITED
7-32 Implementing a Tabular Data Model by Using SQL Server Analysis Services

4. In the Order Date table, hide the DateKey, DayNumberOfWeek, and MonthNumberOfYear
columns, and then rename the columns as shown in the following table:

Column New name

FullDateAlternateKey Date

EnglishDayNameOfWeek Weekday

DayNumberOfMonth Day of Month

EnglishMonthName Month

CalendarQuarter Calendar Quarter

CalendarYear Calendar Year

FiscalQuarter Fiscal Quarter

FiscalYear Fiscal Year

5. In the Ship Date table, hide the DateKey, DayNumberOfWeek, and MonthNumberOfYear
columns, and then rename the columns as shown in the following table:

Column New name

FullDateAlternateKey Date

EnglishDayNameOfWeek Weekday

DayNumberOfMonth Day of Month

EnglishMonthName Month

CalendarQuarter Calendar Quarter

CalendarYear Calendar Year

FiscalQuarter Fiscal Quarter

FiscalYear Fiscal Year

 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 7-33

6. In the Product table, hide the ProductKey, ProductAlternateKey, and ProductSubcategoryKey

columns, and then rename the columns as shown in the following table:

Column New name

EnglishProductName Product Name

StandardCost Standard Cost

ListPrice List Price

SizeRange Size Range

EnglishDescription Description

7. In the Product Subcategory table, hide the ProductSubcategoryKey,

ProductSubcategoryAlternateKey, and ProductCategoryKey columns, and then rename the
EnglishProductSubcategoryName column to Subcategory.
8. In the Product Category table, hide the ProductCategoryKey and ProductCategoryAlternateKey
columns, and then rename the EnglishProductCategoryName column to Category.
9. In the Reseller Sales table, hide all columns other than the Quantity, Cost, and Revenue measures
that you created in the previous exercise.

 Task 3: Configure Column Sort Order

1. In the Ship Date table, configure the sort orders of the following columns:

o The Weekday column should be sorted by the DayNumberOfWeek column.

o The Month column should be sorted by the MonthNumberOfYear column.

2. In the Order Date table, configure the sort orders of the following columns:

o The Weekday column should be sorted by the DayNumberOfWeek column.

o The Month column should be sorted by the MonthNumberOfYear column.

 Task 4: Create Hierarchies

1. In the Geography table, create a hierarchy named Location that includes the following columns:

o Country or Region

o State or Province

o City

o Postal Code
2. In the Order Date table, create a hierarchy named Calendar Date that contains the following fields:

o Calendar Year

o Calendar Quarter
o Month

o Day of Month
 MCT USE ONLY. STUDENT USE PROHIBITED
7-34 Implementing a Tabular Data Model by Using SQL Server Analysis Services

3. In the Order Date table, create a second hierarchy named Fiscal Date that contains the following
fields:

o Fiscal Year

o Fiscal Quarter

o Month
o Day of Month

 Task 5: Test the Model

1. View the data model in Excel, and then verify that the changes that you have made are reflected in
the PivotTable.

2. When you have finished, close Excel without saving the workbook.

Results: After completing this exercise, you should have a tabular data model that includes renamed
columns, custom sort orders, and specific relationships between tables.

Exercise 3: Deploying a Tabular Data Model Project in SQL Server Analysis

Services
Scenario
You have configured the columns, relationships, and hierarchies of your tabular data model. Now you
must improve the processing performance by separating the more recent data and the older data into
partitions. You must then set up a perspective for an analysis team who are interested in breaking down
sales only by product dimensions and not any other dimensions. Next, you must set up a security role for
users who must only have access to data that relates to bike sales in the United States. Finally, you must
deploy the tabular model to the server so that users can connect to it by using client tools.

The main tasks for this exercise are as follows:

1. Separate Data into Partitions

2. Create a Perspective

3. Create a Security Role

4. Deploy the Reseller Sales Project

5. Use the Deployed Tabular Database

 Task 1: Separate Data into Partitions

1. Create a new partition in the Reseller Sales table.
2. Rename the two partitions in the Reseller Sales table to Before 2008 and Since 2008.

3. Set the Before 2008 partition to only show records in the Reseller Sales table that have an order
date before 2008.

4. Set the Since 2008 partition to only show records in the Reseller Sales table that have an order date
that is greater than or equal to January 1, 2008.

5. Use SQL Server Data Tools to process both partitions, using Process Full mode.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 7-35

 Task 2: Create a Perspective

1. Create a perspective called Products Analysis that only shows the Reseller Sales table, the Product
Category table, the Product Subcategory table, and the Products table.

2. View the data model in Excel by using the Products Analysis perspective. Note that only the tables
that you specified for the perspective are visible.

3. When you have finished, close Excel without saving the workbook.

 Task 3: Create a Security Role

1. Create a security role called US Bike Sales, with Read and Process permission.

2. Set a row filter, so that the new role only shows reseller sales where the value of the Category column
in the Product Category table is Bikes, and the value of the Country or Region Code column in the
Geography table is US.

3. View the data model in Excel by using the US Bike Sales role. Note that the reseller data is only
showing where the location is the United States and product category is Bikes.

4. When you have finished, close Excel without saving the workbook.

 Task 4: Deploy the Reseller Sales Project

1. Configure the project for deployment:

o The project should be deployed to the localhost\SQL2 instance of SQL Server Analysis Services.
o The deployed database should be named AdventureWorksTab.

o The deployed model should be named Reseller Sales.

2. Build and deploy the project, and then close SQL Server Data Tools.

 Task 5: Use the Deployed Tabular Database

1. Create a blank workbook in Excel, and then import the data model from SQL Server Analysis Services:

o Use the From Analysis Services source on the Data tab of the ribbon.
o Connect to the MIA-SQL\SQL2 instance of SQL Server Analysis Services by using Windows
authentication.

o Select the Reseller Sales cube in the AdventureWorksTab database.

o Import the data into a PivotTable in the workbook.

2. Explore the data in the PivotTable. When you have finished, close Excel without saving the workbook.

Results: After this exercise, you should have deployed the tabular data model project.
 MCT USE ONLY. STUDENT USE PROHIBITED
7-36 Implementing a Tabular Data Model by Using SQL Server Analysis Services

Module Review and Takeaways

In this module, you have learned how to develop and deploy a tabular data model for SQL Server Analysis
Services by using SQL Server Data Tools.
 MCT USE ONLY. STUDENT USE PROHIBITED
8-1

Module 8
Introduction to Data Analysis Expressions (DAX)
Contents:
Module Overview 8-1 
Lesson 1: DAX Fundamentals 8-2 

Lesson 2: Using DAX to Create Calculated Columns and Measures in a Tabular

Data Model 8-12 

Lab: Using DAX to Enhance a Tabular Data Model 8-24 

Module Review and Takeaways 8-30 

Module Overview
You can extend Microsoft® SQL Server® Analysis Services tabular data models by using the Data Analysis
Expressions (DAX) language. DAX is a highly flexible language that enables you to create measures and
calculated columns for use in PivotTable tables and PivotChart charts.

Objectives
This module explains the fundamentals of the DAX language. It also explains how you can use DAX to
create calculated columns and measures, and how you can use them in your tabular data models.

After completing this module, you will be able to:

 Describe the fundamentals of DAX.

 Use DAX to create calculated columns and measures.
 MCT USE ONLY. STUDENT USE PROHIBITED
8-2 Introduction to Data Analysis Expressions (DAX)

Lesson 1
DAX Fundamentals
To use DAX effectively, you should understand the components of the language and its capabilities. This
lesson explains the fundamentals of DAX and provides an overview of its functionality.

Lesson Objectives
After completing this lesson, you will be able to:

 Describe DAX and how it is used.

 Describe the different types of DAX functions.

 Describe the syntax for writing DAX formulas.

 Describe aggregations.

 Explain why context is important when writing DAX formulas.

 Describe the basics of DAX queries.

Overview of DAX
The DAX language consists of a library of
functions, operators, and constants—you can use
these to create formulas that define the calculated
columns and measures to extend the functionality
of tabular data models. DAX was introduced in the
first version of the PowerPivot for Excel® add-in,
as the language for writing the formulas that
define business logic in PowerPivot for Excel
workbooks. In SQL Server, you can use DAX in
tabular data models in SQL Server Analysis
Services, in addition to those you create in
PowerPivot for Excel workbooks, and in Power BI
Desktop.

Typical Uses of DAX

You can use DAX to extend tabular data models beyond the basic aggregation of numerical columns,
based on tables and hierarchies. With DAX, you can:
 Create columns and measures based on complex custom calculations.

 Create columns that reference data in related tables.

 Define parent-child hierarchies.

 Create Key Performance Indicators (KPIs).

DAX Syntax
DAX is based on the Microsoft Excel formula syntax and uses a range of functions, operators, and values.
However, unlike formulas in Excel, DAX does not work on individual cells or ranges of cells. Instead, you
use it to work with the relational columns and tables that you import into a tabular data model. DAX also
includes more advanced functionality provided by Excel formulas. The similarity between DAX and Excel
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 8-3

formulas minimizes the learning time for information workers who want to learn to use DAX, enabling
faster adoption of the language and reducing the need for training.

For links to the DAX documentation, see Data Analysis Expressions (DAX) Reference in the Microsoft
Developer Network:

Data Analysis Expressions (DAX) Reference

http://aka.ms/Sxj8h6

DAX Functions
DAX provides a range of functions you can use to
build formulas. Many of these functions are the
same as or similar to those used to build formulas
in Excel. DAX, however, includes many additional
functions that help you to create much more
complex and sophisticated reports than you can
by using only Excel functions. For example, DAX
includes functions that can navigate table
relationships, take account of context to calculate
values in PivotTable tables, or compare time
periods to produce the desired result. A DAX
formula can contain multiple functions that can be
nested within other functions (up to 64 levels deep).

Note: You should avoid having too many levels of nesting in your formulas because it
makes them difficult to understand and troubleshoot.

DAX includes several categories of functions, including:

 Text. These functions include CONCATENATE, TRIM, and BLANK. They are based on the Excel string
functions, and enable you to manipulate strings.

 Information. These functions include PATH, ISBLANK, and ISERROR. They enable you to test row
values to see whether they match a type of value, such as an error or a null, and to handle
relationships in a table. This category also includes functions such as USERNAME, which you can use
to implement row-level security in SQL Server Analysis Services tabular data models.

 Filter and value. These functions include CALCULATE, ALL, and FILTER. You can use them to
manipulate data in a variety of ways, such as modifying the context for calculations, or using the
relationships between tables to return specific results. There is no direct equivalent of these functions
in Excel.

 Logical. These functions include TRUE, IF, SWITCH, and NOT. With these functions, you can perform
logical tests on expressions to return information about the state of the tested values, and to use
conditional logic.

 Mathematical and trigonometric. These functions include ROUND, FLOOR, and POWER. They are
similar to the mathematical and trigonometric functions in Excel.

 Statistical and aggregation. These functions include SUM, COUNT, RANKX, and DISTINCTCOUNT.
They are similar to aggregation functions in Excel.
 MCT USE ONLY. STUDENT USE PROHIBITED
8-4 Introduction to Data Analysis Expressions (DAX)

 Date and time. These functions include DATE, MONTH, and YEAR. They are equivalent to date and
time functions in Excel.

 Time intelligence. These functions include PARALLELPERIOD, TOTALYTD, and

OPENINGBALANCEYEAR. You can use these functions to work with columns using the Date data type
in a more sophisticated way; for example, by comparing specific business measures across time
periods.

 Navigation and Lookup. DAX is designed to work with multiple tables and columns in a tabular data
model. It provides functions such as RELATEDTABLE, USERELATIONSHIP, CROSSJOIN, and
LOOKUPVALUE, which you can use to combine data from multiple tables. You can also use functions
such as PATH and PATHITEM to define navigable hierarchies of column values in a single table.

For complete documentation about the functions supported by DAX, see DAX Function Reference in the
Microsoft Developer Network:

DAX Function Reference

http://aka.ms/wqmicc

DAX Syntax and Data Types

DAX is syntactically similar to the Excel formula
language, so users already skilled in this area
should be able to start creating DAX formulas
quickly. The following list summarizes basic DAX
syntax and conventions:
 All DAX formulas must start with an equal sign
(=), followed by an expression.

 Expressions can contain functions, operators,

constants, and references to columns in tables.
Object names that are referenced in
expressions, such as table and column names,
are case insensitive and can include spaces.
You cannot use leading or trailing spaces, characters invalid in the names of SQL Server Analysis
Services objects, or control characters in table, column, or measure names. Arithmetic, comparison,
text concatenation, and logical operators are largely the same as for formulas in Excel, and mostly
work in the same way.

 DAX functions use tables and columns as inputs. To avoid ambiguity, you should use fully qualified
names when referencing columns. A fully qualified name uniquely identifies a column because it
includes the name of the table to which the column belongs. Use square brackets around the column
names and single straight quotation marks around table names that contain spaces; for example,
'Order Date'[Calendar Year]. You can use unqualified names to refer to columns in the same table.
For example, when used in the context of the Order Date table, the 'Order Date'[Calendar Year]
column can be referenced as [Calendar Year].
 When you are writing and referencing one measure to another, you should enclose the measure
name in square brackets; for example, [Total Profit].

Although DAX formulas are similar to Excel formulas, there are two important differences:

 Excel formulas can reference ranges of cells or individual cells. DAX formulas can only act on
complete tables or columns of data. To work with a subset of rows in a table or column, you must use
a filter function to restrict the data.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 8-5

 The data types supported by DAX formulas are not identical to the data types recognized by Excel
formulas.

For more information about DAX syntax, see DAX Syntax Reference in the Microsoft Developer Network:

DAX Syntax Reference

http://aka.ms/Pujroo

Operators
DAX operators are very similar to operators in Excel formulas. The following operators are supported by
DAX:

 Arithmetic:

o +. Addition

o -. Subtraction

o *. Multiplication

o /. Division
o ^. Exponentiation

 Comparison:

o =. Equal
o <. Less than

o >. Greater than

o <=. Less than or equal to

o >=. Greater than or equal to

o <>. Not equal

 Text concatenation:

o &. Concatenate text

 Logical:

o &&. Logical AND

o ||. Logical OR

o IN. Logical OR between rows.

For more information about DAX operators, see DAX Operator Reference in the Microsoft Developer
Network:

DAX Operator Reference

http://aka.ms/Veqptn

Data Types
DAX will implicitly convert data types where possible, so there is no need to explicitly convert a column to
a different data type. For example, if you enter a date as a string and use that value in a date function,
DAX will attempt to convert the data type to Date. There are no DAX functions for explicitly converting
data types.
 MCT USE ONLY. STUDENT USE PROHIBITED
8-6 Introduction to Data Analysis Expressions (DAX)

DAX uses the following data types:

 I8 (eight-byte integer).

 R8 (eight-byte real number).

 Boolean. TRUE or FALSE.

 String.

 Date/time.

 Currency.

 BLANK. Represents a blank or missing value or an empty cell in an Excel worksheet. You can test for
blank values by using the ISBLANK function.

 Table. Used by functions that either require a table as input or return a table.

Note: BLANK and NULL are not the same thing. Handling of BLANK data types varies,
depending upon the operation that you perform. For example, BLANK + 5 = 5, but BLANK * 5 =
BLANK. However, in the SQL Server Database Engine, both NULL + 5 and NULL * 5 result in a
NULL value. Database Engine NULL values are implicitly converted to BLANK values when you
import data into a tabular data model.

Aggregations
Information workers are typically interested in
finding answers to questions such as: "How much
revenue did the company generate though its
Internet sales channel in April?" or: "What was the
best-selling product in France last year?" To
answer these kinds of questions, you must group
and aggregate data. You group data by using
values that give meaning and context, such as
Product Category or Sales Region. These values
are usually stored in the tables you import from
data sources. You can aggregate data by using
DAX aggregation functions. The following are
commonly used aggregation functions:

SUM
Adds up the values in a numeric column, providing a single total for that column.
For example, the following code totals all the values in the Sales Amount column in the Reseller Sales
table:

DAX SUM
=SUM('Reseller Sales'[Sales Amount])
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 8-7

SUMX
Filters and then adds up the values in a numeric column, returning a single total for that filtered column.

For example, the following code returns the total of the Freight column in the Internet Sales table, but
the calculation only includes rows for which the SalesTerritoryKey value is 5:

DAX SUMX
=SUMX(FILTER('Internet Sales', 'Internet Sales'[SalesTerritoryKey]=5),[Freight])

COUNT
Returns a count of the number of rows that are not blank for columns containing either numbers or dates.
COUNTA is similar, but returns a count of the number of rows that are not blank for columns containing
non-numeric values.

For example, the following code returns a count of the rows in the Order Date column in the Reseller
Sales table:

DAX COUNT
=COUNT('Reseller Sales'[Order Date])

COUNTAX
Filters and then counts the number of values that are not blank in a column and returns a single count for
that filtered column. The COUNTX function works in the same way as COUNTAX, but only counts numeric
values. The COUNTBLANK function returns the number of blank values in a column, and the
COUNTROWS function counts the number of rows in a table.

For example, the following code returns the number of rows in the Phone column in the Reseller table,
where the value in the Status column is set to "Active":

DAX COUNTX
=COUNTAX(FILTER('Reseller',[Status]="Active"),[Phone])

MIN and MAX

Return the minimum and maximum values in a numeric column. The MAXA and MINA functions return
the maximum and minimum numeric values in a column, but they also handle non-numeric logical values,
such as TRUE and FALSE, which evaluate to 1 and 0 respectively.

The following code example returns the maximum value for the Sales Amount column in the Reseller Sales
table:

DAX MAX
=MAX('Reseller Sales'[Sales Amount])

MAXX and MINX

Return the maximum and minimum values in a column in a table filtered by using an expression.

For example, the following code returns the minimum value for the Freight column in the Internet Sales
table for which the corresponding SalesTerritoryKey value is 5:

DAX MINX
=MINX(FILTER('Internet Sales', [SalesTerritoryKey] = 5),[Freight])
 MCT USE ONLY. STUDENT USE PROHIBITED
8-8 Introduction to Data Analysis Expressions (DAX)

AVERAGE
Returns the mean of all values in a numeric column. The AVERAGEA function does the same thing as
AVERAGE, but also handles non-numeric data types.

For example, the following code returns the average value in the Total Product Cost column in the
Reseller Sales table:

DAX AVERAGE
=AVERAGE('Reseller Sales'[Total Product Cost])

AVERAGEX
Enables you to supply an expression to obtain the average of values that are not neatly contained in a
single column.

For example, the following code sums the Freight and TaxAmt values for each row in the Internet Sales
table, and then calculates the average value:

DAX AVERAGEX
=AVERAGEX('Internet Sales', 'Internet Sales'[Freight]+ 'Internet Sales'[TaxAmt])

Using Aggregations
When you add a column to the Values area in the PowerPivot Field List in a PowerPivot for Excel
worksheet, column values are automatically summed and the total added to the PivotTable table or
PivotChart chart that you are using. If you need to summarize the data differently, you can select another
type of aggregation, such as COUNT, MIN, or AVERAGE, from a list. The DAX formula is then
automatically created for you. When you need to create aggregations more complex than this, you can
create a measure.
For a complete list of statistical and aggregate functions in DAX, see Statistical Functions (DAX) in the
Microsoft Developer Network:

Statistical Functions (DAX)

http://aka.ms/Ltbpp9

Context
When you create a DAX measure or calculated
column, you define a field you can use in a
PivotTable or a PivotChart. However, the specific
values appearing in any given PivotTable or
PivotChart will depend on the context, such as the
columns and rows added to a PivotTable or a
PivotChart, or the slicers that are enabled at the
time. For example, if you create a PivotTable and
add only the Sum of Sales Amount measure to
the Values area and nothing to the Row Labels
area, you will see a single value representing the
total sales amount for all categories, years, and
countries in the Reseller Sales channel. If you then add Calendar Year to the Row Labels area, you will
see multiple values, each one a total for a given year. Even though the same measure is used in both
cases, the displayed value changes to reflect the context. This context-aware behavior makes it possible to
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 8-9

perform dynamic analysis of data quickly and easily, without needing to reconfigure PivotTables or
PivotCharts whenever you add a new measure.

It is important to be aware of context when you create measures and calculated columns. If you do not
take context into account, you might find your reports do not aggregate data in the way you intended,
and are inaccurate and misleading.

Every DAX expression has two contexts; a row context, and a filter context.

Row Context
The row context of a DAX expression corresponds to a single row in the source data. Row context is used
when you use functions that operate on two or more columns in the same row.

For example, the formula in the following code references the First Name and Last Name columns in the
Employee table:

DAX Row Context

=CONCATENATE([First Name], CONCATENATE(" ", [Last Name]))

The calculated column evaluates the expression for each row. It uses the row context to obtain the values
in the First Name and Last Name columns for each row, and then uses these values to create a new
value for each row by concatenating them. Row context is important when using any function that
evaluates an expression for each row in a table, such as SUMX.

Filter Context
The filter context of a DAX expression is made up of all the filters that are applied to the data when the
expression is evaluated. Filter context is most obviously seen in PivotTables; when you add columns to the
Rows or Columns area of a PivotTable, aggregated columns in the Values are split according to the row
and column values. The row and column values define the filter context for each value cell. PivotTable
slicers also affect the filter context by excluding some values from the PivotTable.

Filter context is important in DAX because you can use DAX functions to modify the row and query
contexts by creating filters in formulas. You can use the FILTER function to produce a subset of rows on
which you can perform a calculation. For example, you might filter by specifying a product category, then
aggregate the sales amount for that category only. You can override filters by using the ALL and
ALLEXCEPT functions. ALL forces all filters to be ignored, and ALLEXCEPT helps you to selectively override
them by specifying particular columns to retain in the filter context—all other columns are ignored.

Query Context
When you add a measure to a PivotTable, the xVelocity engine evaluates the measure value for each cell.
The evaluation takes account of the query context to calculate these values for a given cell. Query context
is determined by the row labels, column labels, filters, and slicers that apply to the PivotTable table.
Removing or adding a new slicer, row label, or column label, will change the filter context; the PowerPivot
engine must take account of the new context that is performing the calculation for each row in the query.
For more information about context, see the Context in DAX Formulas section in Understanding DAX in
Tabular Models (SSAS Tabular) in Microsoft Docs:

Understanding DAX in Tabular Models (SSAS Tabular) - Context in DAX Formulas

http://aka.ms/Jx680b
 MCT USE ONLY. STUDENT USE PROHIBITED
8-10 Introduction to Data Analysis Expressions (DAX)

DAX Queries
DAX includes extensions that help you to write
queries that retrieve data. Client reporting tools,
such as Power View, use DAX to query tabular
data models in SQL Server Analysis Services. It is
unlikely that you will need to write DAX queries
manually, but a basic understanding can help with
troubleshooting.

You can retrieve data from tables by using the

EVALUATE keyword, which is functionally
equivalent to the SELECT statement in SQL Server.

The following code example returns all the rows

from the Reseller Sales table:

Simple EVALUATE
EVALUATE('Reseller Sales')

Queries can include FILTER to return a reduced result set, similar to the way in which a WHERE clause
returns a reduced result set in SQL Server.

The following code example returns all the rows from the Reseller Sales table where the OrderDateKey
value is greater than 20040101:

FILTER and EVALUATE

EVALUATE(FILTER('Reseller Sales', 'Reseller Sales' [OrderDateKey]>20040101))

You can also order result sets.

The following code example returns all the rows from the Reseller Sales table where the OrderDateKey
value is greater than 20040101. The results are ordered by Sales Amount:

FILTER and EVALUATE with ORDER BY

EVALUATE(FILTER('Reseller Sales', 'Reseller Sales'[OrderDateKey]>20040101))
ORDER BY 'Reseller Sales'[Sales Amount] ASC

For more information about DAX queries, see DAX Queries in the Microsoft Developer Network:
DAX Queries
http://aka.ms/Rnt1dc
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 8-11

Check Your Knowledge

Question

When a DAX expression is evaluated,

which contexts apply?

Select the correct answer.

The row context.

The filter context.

Both the row context and the filter

context.
 MCT USE ONLY. STUDENT USE PROHIBITED
8-12 Introduction to Data Analysis Expressions (DAX)

Lesson 2
Using DAX to Create Calculated Columns and Measures in
a Tabular Data Model
DAX creates calculated columns and measures that you can then use in PivotTable tables and PivotChart
charts in PowerPivot for Excel workbooks. This lesson shows you how to create DAX calculated columns
and measures using a variety of functions, including time intelligence functions and functions that
manipulate table relationships. This lesson also shows you how to create KPIs with DAX.

Lesson Objectives
At the end of this lesson, you will be able to:

 Use calculated columns.

 Use measures.
 Use DAX KPIs.

 Work with relationships between tables.

 Understand time intelligence functions.

 Work with hierarchical data.

Calculated Columns
A calculated column is a named column that you
populate by using a DAX formula. The formula will
usually refer to other columns, either in the same
table or in a different, related table, but you can
also create calculated columns based on measures,
or on other calculated columns.
The following code example creates a column that
calculates profit by subtracting the value in the
Total Product Cost column from the value in the
Sales Amount column. You could use this DAX
expression to create a calculated column named
Sales Profit.

A DAX Expression for a Calculated Column

=[Sales Amount] - [Total Product Cost]

You can create calculated columns in a tabular data model just as you can in an Excel workbook. In the
Data View window of a data model designer, select the table to which you want to add the column, and
then click Add Column. You can then name the column and type the DAX formula in the formula bar,
which will determine the contents of the column. The xVelocity engine performs the calculation
immediately to populate the column. By default, the engine recalculates columns automatically if the
source data changes.

Note: The formula used to create a calculated column applies to every row. You cannot
apply the formula only to selected data ranges, as you can in Excel worksheets.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 8-13

When creating calculated columns, be aware of the following considerations:

 Calculated columns calculate a value for every row in a table. This can take a long time when tables
contain a large number of rows.

 Changing the names of columns referenced by a calculated column will cause an error.

For more information about DAX calculated columns, see Calculated Columns in Microsoft Docs:
Calculated Columns
http://aka.ms/She68f

Measures
A measure is a named formula that can
encapsulate complex business logic, usually by
aggregating one or more numerical values.
The following example defines a measure named
Profit that calculates the sum of the Sales Profit
column:

A DAX Expression for a Measure

Profit:=SUM([Sales Profit])

Measures are defined in the Measures Grid, in

Data View, of the data model designer. In client
tools, measures are associated with the table in which they have been defined. If you need to create a
measure that is not associated with an existing table, you can paste an Excel worksheet containing only a
placeholder column header value into the data model as a new, empty table. You can then define
measures in the measure grid for the empty table.

For more information about DAX measures, see Measures in Microsoft Docs:

Measures
http://aka.ms/K55jub

KPIs
KPIs are used to compare measures with a given
goal or target value to assess business
performance. For example, a business might set a
profit margin goal of 15 percent, and use a KPI to
indicate how actual sales results compare.
In tabular data models, KPIs are defined based on
measures, and can consist of the following
elements:
 Base Value: the actual value being assessed,
calculated by a measure.

 Target Value: the goal against which the base

value is compared. This can be an absolute value or calculated by a measure.
 MCT USE ONLY. STUDENT USE PROHIBITED
8-14 Introduction to Data Analysis Expressions (DAX)

 Status Thresholds: the status is an indicator of the base value’s performance against the target value.
You can set thresholds that establish the status, based on the percentage of the target value achieved
by the base value. For example, you could use a red indicator if the base value is below 20 percent of
the target value; a yellow indicator if it is between 20 and 80 percent; and a green indicator if it is
above 80 percent.
For more information about working with DAX KPIs, see KPIs in Microsoft Docs:

KPIs
http://aka.ms/Cle5ry

Demonstration: Creating and Using Measures and Calculated Columns

In this demonstration, you will see how to:

 Create calculated columns.

 Create measures.
 Create a KPI.

 Analyze calculated columns and measures.

Demonstration Steps
Create Calculated Columns

1. Ensure that the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and log
on to the 20768C-MIA-SQL virtual machine as ADVENTUREWORKS\Student with the password
Pa55w.rd.

2. In the D:\Demofiles\Mod08 folder, right-click Setup.cmd and click Run as administrator. When
prompted, click Yes.

3. Start Visual Studio 2015 and open the TabularDemo.sln solution in the D:\Demofiles\Mod08
folder.
4. If the Tabular model designer dialog box appears, in the Workspace server list, specify
localhost\SQL2 and then click OK.

5. If the Tabular Model Designer dialog box appears, click Yes.

6. In Solution Explorer, double-click Model.bim to open the data model designer.

7. On the Model menu, point to Process, and click Process All. If you are prompted for impersonation
credentials, specify the user name ADVENTUREWORKS\ServiceAcct with the password Pa55w.rd
and click OK.
8. In the Data Processing dialog box, when all of the tables have been processed, click Close.

9. On the Customer tab, in the first empty column after the existing columns, double-click Add
Column and name the new column Full Name.

10. With the Full Name column selected, in the formula bar, enter the following DAX expression:

=CONCATENATE([FirstName], " " & [LastName])

Wait for the table to finish updating, and note the calculated values in the new column.

11. Click the FirstName column heading, hold Shift and click the LastName column heading, and then
right-click either of the selected column headings and click Hide from Client Tools.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 8-15

12. On the Internet Sales tab, in the first empty column after the existing columns, double-click Add
Column and name the new column SalesProfit.

13. With the SalesProfit column selected, in the formula bar, enter the following DAX expression:

=[SalesAmount]-[TotalProductCost]

Wait for the table to finish updating, and note the calculated values in the new column.

Create Measures

1. On the Internet Sales tab, click the first empty cell in the measure grid under the SalesProfit
column.

2. In the formula bar, enter the following DAX expression:

Profit:=SUM([SalesProfit])

3. Select the cell containing the Profit measure, and press F4 to display the properties pane. Then set the
Format property to Currency.

4. Click the empty cell in the measure grid under the Profit measure.

5. In the formula bar, enter the following DAX expression:

Margin:=[Profit]/[Revenue]

6. Select the cell containing the Margin measure, and press F4 to display the properties pane. Then set
the Format property to Percentage.

7. Right-click the SalesProfit column header and click Hide from Client Tools.

8. On the File menu, click Save All.

Create a KPI

1. On the Internet Sales tab, click the empty cell in the measure grid under the Margin measure.

2. In the formula bar, enter the following DAX expression:

Target Margin:=(SUM('Product'[ListPrice]) - SUM('Product'[StandardCost])) /

SUM('Product'[ListPrice])

3. Select the cell containing the Margin measure, and press F4 to display the properties pane. Then set
the Format property to Percentage.
4. On the Internet Sales tab, right-click the cell in the measure grid containing the Margin measure
and click Create KPI.

5. In the Key Performance Indicator (KPI) dialog box, note that the KPI base measure (value) is
defined by the Margin measure. Then, under Target, ensure that Measure is selected and select
Target Margin.

6. Set the first status threshold to 65% and the second to 90%.

7. Note the default icon style, and click OK.

8. On the File menu, click Save All.

 MCT USE ONLY. STUDENT USE PROHIBITED
8-16 Introduction to Data Analysis Expressions (DAX)

Analyze Columns and Measures

1. On the Model menu, click Analyze in Excel.

2. In the Analyze in Excel dialog box, ensure that Current Windows User is selected and that the
(Default) perspective is specified, and click OK.
3. In Excel, in the PivotTable Fields pane, in the Internet Sales table, select the Profit and Margin
measures so that they are summarized in the PivotTable.

4. In the PivotTable Fields pane, in the Customer table, select Full Name so that the PivotTable shows
profit and margin by customer.

5. In the PivotTable Fields pane, expand KPIs, expand Margin, and select Status so that an indicator
shows the KPI status for each customer.

6. Close Excel without saving the workbook.

7. Keep Visual Studio open for the next demonstration.

Multiple Relationships
Most data models consist of multiple tables that
are related, based on common key values. There
are many cases where a DAX expression needs to
combine data from related tables, and DAX
provides some functions to support this
requirement.

Looking Up Related Data Values

DAX provides a LOOKUPVALUE function you can
use to return a column value from a table, based
on a specified value for another column.

For example, if a Customer table contains a

Registration Date column, and a Date table contains Full Date and Fiscal Year columns, you could use
the following expression in a calculated column in the Customer table to look up the fiscal year in which
the customer registered:

DAX LOOKPVALUE
LOOKUPVALUE('Date'[Fiscal Year], [Full Date], [Registration Date])

The LOOKUPVALUE does not require an existing relationship between tables; it simply looks up a column
value based on the specified value for another column in the same table.
If a relationship is defined between two tables, a more effective way to look up related values is to use the
RELATED function.

For example, if a Product table is related to a Product Category table, based on a common
CategoryKey column, you could use the following expression to return the Category Name value from
the Category table into a calculated column in the Product table:

DAX RELATED
RELATED('Product Category'[Category Name])
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 8-17

Note that the RELATED function does not require that you specify a lookup value. It automatically uses
the active relationship between the tables to find the related value. The RELATED function works well
when you need to look up a single value from the “many” side of a one-to-many relationship. In the
example above, each category exists once in the Product Category table but many times in the Product
table.
For more information about the RELATED function, see RELATED Function (DAX) in the Microsoft
Developer Network:

RELATED Function (DAX)

http://aka.ms/Vram4o

Occasionally, you might want to aggregate many related values from the “one” side of the relationship.

For example, in the Product Category table, you might want to create a calculated column named
Product Count containing the total number of products in each category. To accomplish this task, you
can use the RELATEDTABLE function shown in the following example:

DAX RELATEDTABLE
COUNTX(RELATEDTABLE('Product'), [ProductKey])

For more information about the RELATEDTABLE function, see RELATEDTABLE Function (DAX) in the
Microsoft Developer Network:

RELATEDTABLE Function (DAX)

http://aka.ms/Ey4h2a

Referencing Tables with Multiple Relationships

Tabular data models support multiple relationships between tables. This is useful if, for example, the Date
and Sales tables have multiple relationships to support different dates, such as order date, due date, and
delivery date. Multiple relationships appear in the diagram view of the data model designer, with the
active relationship shown as an unbroken line. Non-active relationships appear as dotted lines.
Only one relationship can be active and this will automatically be used when performing calculations
using data from the two tables. You can use the USERELATIONSHIP function when you are writing DAX
formulas to override this behavior and force PowerPivot to use the relationship that you specify. For
example, if the default relationship between your fact table and date table is based on the Order Date
column—but you need to aggregate data by shipping date—you can write a DAX measure containing the
USERELATIONSHIP function to achieve this.

Note: USERELATIONSHIP does not create a relationship. The relationship must already exist
for you to use it.

For example, the following code calculates the sum of the Sales Amount column based on shipping date:

DAX USERELATIONSHIP
=CALCULATE(SUM(Reseller Sales[Sales Amount]), USERELATIONSHIP('Reseller
Sales'[ShipDateKey], 'Date'[DateKey]))

Note: You can only use USERELATIONSHIP with other functions that accept a filter as an
argument, such as CALCULATE in the previous example.
 MCT USE ONLY. STUDENT USE PROHIBITED
8-18 Introduction to Data Analysis Expressions (DAX)

For more information about the USERELATIONSHIP function, see USERELATIONSHIP Function (DAX) in the
Microsoft Developer Network:

USERELATIONSHIP Function (DAX)

http://aka.ms/Vdbozm

For more information about working with relationships in tabular models, see Relationships (SSAS Tabular)
in Microsoft Docs:

Relationships (SSAS Tabular)

http://aka.ms/Guu6au

Demonstration: Using a Specific Relationship in a Measure

In this demonstration, you will see how to:
 Retrieve a value from an unrelated table.

 Retrieve a value from a related table.

 Aggregate related values.

 Analyze related data.

Demonstration Steps
Retrieve a Value from an Unrelated Table
1. Ensure that you have completed the previous demonstration.

2. In Visual Studio, with the Model.bim pane open in data view, on the Customer tab, after the existing
columns, double-click the Add Column header and enter the column name First Fiscal Year.
3. With the First Fiscal Year column selected, in the formula bar, enter the following DAX expression:

=LOOKUPVALUE('Order Date'[FiscalYear], [Date], [DateFirstPurchase])

Wait for the table to finish updating, and note the calculated values in the new column.

Retrieve a Value from a Related Table

1. On the Customer tab, after the existing columns, double-click the Add Column header and enter the
column name Country-Region.

2. With the Country-Region column selected, in the formula bar, enter the following DAX expression:

=RELATED('Geography'[CountryRegionCode])

Wait for the table to finish updating, and note the calculated values in the new column.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 8-19

Aggregate Related Values

1. On the Customer tab, after the existing columns, double-click the Add Column header and enter the
column name Order Count.

2. With the Order Count column selected, in the formula bar, enter the following DAX expression:

=COUNTAX(RELATEDTABLE('Internet Sales'), [SalesOrderNumber])

Wait for the table to finish updating, and note the calculated values in the new column.

3. On the File menu, click Save All.

Analyze Related Data

1. On the Model menu, click Analyze in Excel.

2. In the Analyze in Excel dialog box, ensure that Current Windows User is selected and that the
(Default) perspective is specified, and click OK.

3. In Excel, in the PivotTable Fields pane, in the Internet Sales table, select the Revenue measure.

4. In the PivotTable Fields pane, in the Customer table, select First Fiscal Year so that the PivotTable
shows sales revenue by customer, based on the first year in which customers made a purchase.

5. In the PivotTable Fields pane, in the Customer table, clear First Fiscal Year and select Country-
Region so that the PivotTable shows sales revenue by customer country or region.

6. In the PivotTable Fields pane, in the Customer table, clear Country-Region and select Order
Count so that the PivotTable shows sales revenue by customer, based on the number of orders
placed.

7. Close Excel without saving the workbook.

8. Leave Visual Studio open for the next demonstration.

Introduction to Time Intelligence

You can use time intelligence functions to
compare data from a time period against
equivalent data from another time period. For
example, you might want to compare sales for the
first quarter of this year against sales for the first
quarter of last year.
To support time intelligence functionality, you
should ensure that your data and tables meet the
following criteria:
 The model should have a separate table that
contains only date information.

 The date table should have a continuous range of dates without gaps.

 The date table column that uses the date data type should use “day” as the lowest level of granularity.

Relationships between a table containing a measure and a date table do not need to be based on
columns that use the date data type to support time intelligence functionality. Instead, you can base these
relationships on columns containing integer values. This is useful because it is likely that the databases
you use as sources for analysis will contain fact and date tables that are related by columns using the
 MCT USE ONLY. STUDENT USE PROHIBITED
8-20 Introduction to Data Analysis Expressions (DAX)

integer data type—particularly if the source database is a data warehouse. For example, in the
AdventureWorksDW database, the FactResellerSales table relates to the DimDate table on the
DateKey column. To handle time intelligence correctly when integers are used in relationships, you must
manually mark the date table as Date Table and specify the column in the table that contains the date
values. This column must use the date data type.

Time Intelligence Functions

DAX provides a range of time intelligence functions that support common business analysis scenarios.
These include (among many others):

 NEXTMONTH, NEXTQUARTER, NEXTYEAR. These functions return a table that contains a range of
dates representing the time period after the current context.

 PREVIOUSMONTH, PREVIOUSQUARTER, PREVIOUSYEAR. These functions return a table that contains

a range of dates representing the time period preceding the current context.

 SAMEPERIODLASTYEAR, PARALLELPERIOD. You can use these functions to compare measure values
across the same period in different timespans; for example, to compare sales for a month in the
current year with sales in the same month in the previous year.

 TOTALMTD, TOTALQTD, TOTALYTD. These functions calculate a running total for the corresponding
time period. For example, you could use TOTALQTD to calculate sales for the quarter so far.

For more information about DAX time intelligence functions, see Time Intelligence Functions (DAX) in the
Microsoft Developer Network:

Time Intelligence Functions (DAX)

http://aka.ms/Vwsifx

Demonstration: Using Time Intelligence

In this demonstration, you will see how to:

 Mark a table as a date table.

 Create a measure that uses time intelligence.

 Analyze data using time intelligence.

Demonstration Steps
Mark a Table As a Date Table

1. Ensure you have completed the previous demonstration.

2. In Visual Studio, with the Model.bim pane open in data view, click the Order Date tab.
3. On the Table menu, point to Date and click the Mark As Date Table.

4. In the Mark As Date Table dialog box, select the Date column, and click OK.

Create a Measure That Uses Time Intelligence

1. In the model designer, click the Internet Sales tab.

2. Click the first empty cell in the measure grid, then in the formula bar, enter the following DAX
expression:

Previous Year Revenue:=CALCULATE(SUM([SalesAmount]), SAMEPERIODLASTYEAR('Order

Date'[Date]))
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 8-21

3. Select the cell containing the Previous Year Revenue measure, and press F4 to display the properties
pane. Then set the Format property to Currency.

4. On the File menu, click Save All.

Analyze Data Using Time Intelligence

1. On the Model menu, click Analyze in Excel.

2. In the Analyze in Excel dialog box, ensure that Current Windows User is selected and that the
(Default) perspective is specified, and click OK.
3. In Excel opens, in the PivotTable Fields pane, in the Internet Sales table, select the Revenue and
Previous Year Revenue measures.

4. In the PivotTable Fields pane, in the Order Date table, drag Calendar Date to the Rows area so
that the PivotTable shows sales revenue and previous year revenue for each year.

5. Expand 2007 and 2008, and note that the revenue for each quarter in 2007 is shown as the previous
year revenue for the same quarters in 2008.

6. Close Excel without saving the workbook.

7. Close Visual Studio.

Implementing Parent-Child Hierarchies

Unlike in MDX, there is no inherent support for a
parent-child hierarchy within a tabular data
model. To define a parent-child hierarchy, you
must use DAX expressions including functions that
can be used to specify the path of parent and
child IDs, and the distinct levels you want to
support within the hierarchy. Understanding the
functions available is important because parent-
child hierarchies can be commonplace within an
analytical data model. To create a parent-child
hierarchy in a tabular data model, you must use
the PATH, LOOKUPVALUE, and PATHITEM DAX
functions.

The PATH Function

The PATH function returns a value to denote the entire parent values related to the key value on which
the PATH statement is based. You can create a calculated measure that stores the value of the path item.
For example, you could create an additional column named EmployeeLevels that stores this value in an
Employee table.
The PATH function has the following syntax:

DAX PATH Syntax

PATH(<id_columnName>, <parent_columnName>)
 MCT USE ONLY. STUDENT USE PROHIBITED
8-22 Introduction to Data Analysis Expressions (DAX)

PATH takes the following parameters:

 <id_columnName> refers to the key column in a table. For example, in an Employees table, the key
column might be EmployeeID.

 <parent_columnName> refers to the parent column for a key column in a table. For example, in the
Employees table, this might be ManagerID.

There are other DAX functions similar to PATHITEM that are used to work with other aspects of
hierarchical data, including PATHITEMREVERSE, PATHLENGTH, and PATHCONTAINS. For more
information about all the DAX hierarchy functions, see Parent and Child Functions (DAX) in the Microsoft
Developer Network:

Parent and Child Functions (DAX)

http://aka.ms/Jqr010

The LOOKUPVALUE and PATHITEM Functions

To return a specific value and more meaningful information within a parent-child hierarchy from the
PATH function, you can use the LOOKUPVALUE and PATHITEM functions together. The LOOKUPVALUE
function returns the value to display and PATHITEM determines the level from which the value should be
returned. For example, instead of returning an EmployeeID and a ManagerID in a parent-child hierarchy,
both functions could be used together to return the Name attribute of an employee.

The LOOKUPVALUE function has the following syntax:

DAX LOOKUPVALUE Syntax

LOOKUPVALUE( <result_columnName>, <search_columnName>, <search_value>[,
<search_columnName>, <search_value>]…)

LOOKUPVALUE takes the following parameters:

 <result_columnName> is the name of an existing column that contains the value you want to return.

 <search_columnName> is the name of an existing column on which the lookup is performed.

 <search_value> provides a filter for the LOOKUPVALUE function; this can include a string literal value
or another function, such as PATHITEM, to filter the data.

 Optionally, additional <search_columnName> and <search_value> parameters can be defined.

The PATHITEM function has the following syntax:

DAX PATHITEM Syntax

PATHITEM(<path>, <position>[, <type>])

PATHITEM takes the following parameters:

 <path> refers to a column that contains the results of the PATH function. In the example of an
Employees table, this could be a column named EmployeeLevels.

 <position> is an integer expression referring to the position of the item to be returned.

 <type> is an optional parameter that can be used to determine the data type that the result should
be returned in. A value of 0 is text—which is the default—and a value of 1 is an integer data type.

You can use the LOOKUPVALUE and PATHITEM functions together to create a parent-child hierarchy.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 8-23

An additional calculated measure can be created in a calculated column that contains the following code
to populate the Name value for an employee’s immediate manager:

Employee’s Immediate Manager

LOOKUPVALUE ([Name], [EmployeeId], PATHITEM ([EmployeeLevels], 1))

The calculated measure to create the second level of the employee hierarchy in a second calculated
column is shown in the following code example:

Second Level Manager

LOOKUPVALUE ([Name], [EmployeeId], PATHITEM ([EmployeeLevels], 2))

Design Considerations
When you design parent-child hierarchies in analytical data models, consider the following points:

 Ensure that the parent and child keys are of compatible data types.

 For the best query and processing performance, ensure that a self-join relationship exists between the
parent key and the child.
 Because the manager levels are hard coded, you may need to change the data model if the
organizational structure changes—to allow more levels of management than the model defines.
For more information about working with parent-child hierarchies, see Understanding Functions for
Parent-Child Hierarchies in DAX in the Microsoft Developer Network:

Understanding Functions for Parent-Child Hierarchies in DAX

http://aka.ms/Dqba35

Check Your Knowledge

Question

You are creating a calculated column in a

table, based on the values in a column in
another table. The two tables have a one-
to-many relationship, based on a key
column. Which DAX function should you
use to perform the lookup in the most
efficient way?

Select the correct answer.

USERELATIONSHIP

RELATEDTABLE

RELATED

LOOKUPVALUE
 MCT USE ONLY. STUDENT USE PROHIBITED
8-24 Introduction to Data Analysis Expressions (DAX)

Lab: Using DAX to Enhance a Tabular Data Model

Scenario
A senior business analyst at Adventure Works Cycles wants to perform some detailed analysis of the data
in a tabular data model. To accomplish this, you will need to use DAX to enhance the data model.

Objectives
At the end of this lab, you will be able to:

 Use DAX to create calculated columns.

 Use DAX to create measures.

 Create a KPI that uses a DAX expression for the target value.

 Use DAX to implement a parent-child hierarchy.

Estimated Time: 75 minutes

Virtual machine: 20768C-MIA-SQL

User name: ADVENTURWORKS\Student

Password: Pa55w.rd

Exercise 1: Creating Calculated Columns

Scenario
You have the raw data that you require, but this data would be more useful if it could be adapted by
using calculations.

To improve the presentation of data, you want to concatenate your employees’ first name, middle name,
and last name fields. You want to make profit more visible by creating a calculated value that subtracts a
product’s cost from its selling price. Furthermore, you want to create a calculation to improve the
categorization of products.

The main tasks for this exercise are as follows:

1. Prepare the Lab Environment

2. Concatenate Text Values

3. Calculate a Numeric Value

4. Show Related Values

5. View Calculated Columns in Excel

 Task 1: Prepare the Lab Environment

1. Ensure the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and then log
on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd.

2. Run Setup.cmd in the D:\Labfiles\Lab08\Starter folder as Administrator.

 Task 2: Concatenate Text Values

1. Use Visual Studio to open the AWSalesTab.sln solution in the D:\Labfiles\Lab08\Starter folder, and
view the data model designer. If the Tabular model designer dialog box is displayed, in the
Workspace server list, select localhost\SQL2, and then click OK.

2. View the model, running a script in the server if prompted, then process the model using the
impersonation credentials ADVENTUREWORKS\ServiceAcct with the password Pa55w.rd.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 8-25

3. In the Employee table, create a custom column named Employee Name that concatenates the first
name, middle name (if there is one), and last name of the employee. You can use the following DAX
expression to do this:

=[First Name] & IF(ISBLANK([Middle Name]), "", CONCATENATE(" ", [Middle Name])) &
CONCATENATE(" ", [Last Name])

4. After you have defined the Employee Name column, hide the First Name, Middle Name, and Last
Name columns from client tools and save the data model.

 Task 3: Calculate a Numeric Value

1. In the Reseller Sales table, create a custom column named SalesProfit that calculates profit by
subtracting the total product cost from the sales amount. You can use the following DAX expression
to do this:

=[SalesAmount]-[TotalProductCost]

2. After you have defined the SalesProfit column, confirm that its data type has been automatically
detected as Currency, and then save the model.

 Task 4: Show Related Values

1. In the Product table, create a column named Subcategory that returns the related Subcategory
column from the Product Subcategory table. You can use the following DAX expression to do this:

=RELATED('Product Subcategory'[Subcategory])

Note: Some products at the beginning of the table are uncategorized.

2. In the Product table, create a column named Category that returns the related Category column
from the Product Category table. You can use the following DAX expression to do this:

=RELATED('Product Category'[Category])

Note: Some products at the beginning of the table are uncategorized.

3. In the Product table, create a hierarchy named Categorized Products that includes the following
columns:

o Category

o Subcategory

o Product Name

4. Hide all columns in the Product Subcategory and Product Category tables from client tools.

5. Save the data model.

 Task 5: View Calculated Columns in Excel

1. Analyze the model you have created in Excel; view the Revenue measure by the Categorized
Products hierarchy on rows, and the Employee Name column on columns.

2. Verify that the calculated columns work as expected, and then close Excel without saving the
workbook. Leave Visual Studio open for the next exercise.
 MCT USE ONLY. STUDENT USE PROHIBITED
8-26 Introduction to Data Analysis Expressions (DAX)

Results: After this exercise, you should have a calculated column named Employee Name in the
Employee table, a calculated measure named SalesProfit in the Reseller Sales table, and a hierarchy
named Categorized Products in the Product table.

Exercise 2: Creating Measures

Scenario
Business users have requested the ability to view the following aggregated measures:

 Profit

 Margin
 Previous Year Revenue

The main tasks for this exercise are as follows:

1. Create a Measure That Aggregates a Column

2. Create a Measure That References Other Measures

3. Create a Measure That Uses Time Intelligence

4. View Measures in Excel

 Task 1: Create a Measure That Aggregates a Column

1. In the Reseller Sales table, create a measure named Profit that sums the values in the SalesProfit
column. You can use the following DAX expression to do this:

Profit:=SUM([SalesProfit])

2. Format the Profit measure as Currency.

3. Hide the SalesProfit column from client tools.

4. Save the data model.

 Task 2: Create a Measure That References Other Measures

1. In the Reseller Sales table, create a measure named Margin that divides the Profit measure by
Revenue. You can use the following DAX expression to do this:

Margin:=[Profit]/[Revenue]

2. Format the Margin measure as Percentage.

3. Save the data model.

 Task 3: Create a Measure That Uses Time Intelligence

1. Mark the Order Date table as a date table, specifying the Date column as the table’s unique
identifier.
2. In the Reseller Sales table, create a measure named Previous Year Revenue that uses the
SAMEPERIODLASTYEAR function to calculate the sum of the SalesAmount column for the equivalent
time period in the previous year. You can use the following DAX expression to do this:

Previous Year Revenue:=CALCULATE([Revenue], SAMEPERIODLASTYEAR('Order Date'[Date]))

3. Format the Previous Year Revenue measure as Currency.

4. Save the data model.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 8-27

 Task 4: View Measures in Excel

1. Analyze the model you have created in Excel; view the Revenue, Profit, Margin, and Previous Year
Revenue measures by the Calendar Date hierarchy on rows.

2. Verify that the measures you created all work as expected, and then close Excel without saving the
workbook. Leave Visual Studio open for the next exercise.

Results: At the end of this exercise, the Reseller Sales table should contain the following measures:

Profit
Margin

Previous Year Revenue

Exercise 3: Creating a KPI

Scenario
Adventure Works wants to achieve year-on-year revenue growth of 20 percent. To track this goal, you
plan to create a KPI.

The main tasks for this exercise are as follows:

1. Create a Measure to Calculate a KPI Goal

2. Create a KPI

3. View a KPI in Excel

 Task 1: Create a Measure to Calculate a KPI Goal

1. In the Reseller Sales table, create a measure named Revenue Goal that multiplies the Previous Year
Revenue measure by 1.2. You can use the following DAX expression to do this:

Revenue Goal:=[Previous Year Revenue] * 1.2

2. Format the Revenue Goal measure as Currency and hide it from client tools.
3. Save the data model.

 Task 2: Create a KPI

1. In the Reseller Sales table, create a KPI based on the Revenue measure.

2. Configure the KPI to use the Revenue Goal measure as the target value, and set status thresholds at
75% and 95%.

3. When you have created the KPI, save the data model.

 Task 3: View a KPI in Excel

1. Analyze the model you have created in Excel; view the Value, Goal, and Status of the Revenue KPI
by the Fiscal Date hierarchy on rows.

2. Verify that the KPI indicates the status appropriately for the quarters and months in 2008, and then
close Excel without saving the workbook. Leave Visual Studio open for the next exercise.

Results: At the end of this exercise, the Reseller Sales table should include a measure named Revenue
Goal and a KPI based on the Revenue Goal measure.
 MCT USE ONLY. STUDENT USE PROHIBITED
8-28 Introduction to Data Analysis Expressions (DAX)

Exercise 4: Implementing a Parent-Child Hierarchy

Scenario
The CEO at Adventure Works wants to view sales results aggregated by each sales manager, and then drill
down to view individual employee sales results.

The main tasks for this exercise are as follows:

1. Create a Path Column

2. Create a Column for Each Hierarchy Level

3. Use the Calculated Columns in a Hierarchy

4. View a Parent-Child Hierarchy in Excel

 Task 1: Create a Path Column

1. In the Employee table, add a column named Path that uses the PATH function to define the
recursive chain of parent employee keys. You can use the following DAX expression to do this:

=PATH([EmployeeKey], [ParentEmployeeKey])

2. Save the data model.

 Task 2: Create a Column for Each Hierarchy Level

1. In the Employee table, add the following calculated columns:
o Level1

=LOOKUPVALUE ([Employee Name], [EmployeeKey], PATHITEM ([Path], 1, 1))

o Level2

=LOOKUPVALUE ([Employee Name], [EmployeeKey], PATHITEM ([Path], 2, 1))

o Level3

=LOOKUPVALUE ([Employee Name], [EmployeeKey], PATHITEM ([Path], 3, 1))

o Level4

=LOOKUPVALUE ([Employee Name], [EmployeeKey], PATHITEM ([Path], 4, 1))

2. Save the data model.

 Task 3: Use the Calculated Columns in a Hierarchy

1. Create a hierarchy named Employee Hierarchy in the Employee table.
2. Add the Level1, Level2, Level3, and Level4 columns to the hierarchy, in that order.

 Task 4: View a Parent-Child Hierarchy in Excel

1. Analyze the model you have created in Excel; view the Revenue measure by Employee Hierarchy.

2. Expand Employee Hierarchy to view sales totals for managers and their subordinates. Note that the
personal revenue for sales managers is shown with a blank employee name under the total for that
sales manager.

3. When you have finished, close Excel without saving the workbook, and close Visual Studio.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 8-29

Results: At the end of this exercise, the Employee table should include a hierarchy named Employee
Hierarchy.

Check Your Knowledge

Question

Which DAX time intelligence function

would you use to calculate a measure
based on values for the end of the next
quarter?

Select the correct answer.

ENDOFQUARTER

NEXTQUARTER

OPENINGBALANCEQUARTER

PREVIOUSQUARTER
 MCT USE ONLY. STUDENT USE PROHIBITED
8-30 Introduction to Data Analysis Expressions (DAX)

Module Review and Takeaways

In this module, you have learned how to use DAX to enhance tabular data models.

Review Question(s)
Question: Will time intelligence functionality be useful to you? What kinds of time-based
analyses might you want to perform? Which are the relevant time intelligence functions that
you might use to achieve this?
 MCT USE ONLY. STUDENT USE PROHIBITED
9-1

Module 9
Performing Predictive Analysis with Data Mining
Contents:
Module Overview 9-1 
Lesson 1: Overview of Data Mining 9-2 

Lesson 2: Creating a Custom Data Mining Solution 9-7 

Lesson 3: Validating a Data Mining Model 9-15 

Lesson 4: Connecting to and Consuming a Data Mining Model 9-20 

Lab: Using Data Mining to Support a Marketing Campaign 9-23 

Module Review and Takeaways 9-28 

Module Overview
With data mining, you can use the data you own to gain insights that could help you make intelligent
decisions about your business. Microsoft® SQL Server® Analysis Services includes data mining tools that
you can use to identify patterns in your data, helping you to determine why particular things happen and
to predict what might occur in the future. This module introduces data mining, describes how to create a
data mining solution; how to validate data mining models; and how to incorporate data mining results
into Reporting Services reports.

Objectives
After completing this module, you will be able to:

 Describe the key data mining concepts.

 Create a data mining solution.

 Validate data mining models.

 Use data mining data in a report.

 MCT USE ONLY. STUDENT USE PROHIBITED
9-2 Performing Predictive Analysis with Data Mining

Lesson 1
Overview of Data Mining
Data mining is a type of data analysis that involves using statistical models to reveal connections and
correlations in large sets of data that would otherwise be very difficult, or even impossible, to identify.

This lesson explains the purpose of data mining, and describes the components of a data mining solution,
in addition to the algorithms used to build prediction models.

Lesson Objectives
After completing this lesson, you will be able to:

 Describe the purpose of data mining.

 Describe the components of an Analysis Services data mining solution.

 Describe the different types of data mining algorithms you can use in Analysis Services data mining
solutions.

What Is Data Mining?

Exploration of the data in a data warehouse has
the potential to reveal patterns and trends that
can be useful to organizations in many ways.
These include being able to predict customer
behavior, or to provide customers with product
recommendations in real time as they shop online.
However, the size and complexity of data in large
data warehouses can make it very difficult to
derive useful information from raw data by using
standard data analysis tools and techniques. Data
mining is the statistical analysis of large volumes
of data that would be very difficult to analyze
manually. With data mining, business users can overcome the problems associated with analyzing large
and complex data sets to access useful and actionable information.

Data Mining Algorithms

Data mining involves using algorithms to search through data to extract patterns and trends. For example,
a retail organization might want to discover if customers in large urban areas are more likely to buy goods
in a high price range, compared to customers living in more rural locations. The organization could use
data mining to determine any correlations between location and price; how strong the relationships might
be; or how multiple correlating factors affect outcomes when they are considered together. There are
various data mining algorithms you can use, depending on the type of questions that you need to be
answered.

Data mining can help you to identify trends and patterns that may not be immediately obvious. You can
use data mining to predict unknown values, based on statistics and patterns in previously analyzed sets of
data. This is useful when you are trying to predict and plan for future events.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 9-3

Data Mining Scenarios

Examples of where businesses find value in data mining include:

 Forecasting. Companies can use data mining to analyze sales patterns to determine, for example,
when a particular product will sell or how stores will profit over time.

 Targeted advertising. Organizations can create target marketing and advertising campaigns by
identifying the factors that best predict the customers who are most likely to purchase a given
product, and then targeting the campaign at those individuals.

 Recommendations. Companies can recommend merchandise to customers as they shop online by

analyzing their purchasing histories and previous sales of products.

 Risk assessment. Insurance companies can assess the likelihood of a claim being fraudulent by
employing data mining algorithms that use the outcomes of previous claims to create a weighting for
each factor affecting the new claim. Credit ratings agencies can use financial and customer history
data to predict which individuals are most likely to default on a loan.

For a general introduction to data mining on SQL Server, see Data Mining Concepts in Microsoft Docs:

Data Mining Concepts

http://aka.ms/dz1swn

Data Mining Concepts

The basic unit of data in data mining is a case.
Data mining involves processing cases through an
analytical model. To begin working with data
mining, you must identify one or more data
sources from where your cases will be drawn. You
then create a data mining structure—which
defines cases through metadata—and one or
more data mining models, which hold statistical
patterns found in the case data.

Note: To create a data mining solution in

SQL Server Analysis Services, you must install
Analysis Services in Multidimensional and Data Mining mode. Tabular mode and PowerPivot for
Microsoft SharePoint® mode do not support data mining. You can install multiple instances of
Analysis Services, running in different modes, on the same server.

Data Sources
The source data analyzed by a data mining solution is not stored in the solution—instead, the solution
contains bindings to the data sources. When the data mining solution is processed, an aggregated
summary of the source data is generated; the aggregations may be discarded after processing, or stored
in the solution for use in later operations.

You can use an Analysis Services data source view (DSV) to provide your source data definition; a DSV may
contain bindings to, and define relationships between, multiple data sources. Data sources are not
restricted to OLAP and relational data in SQL Server; Analysis Services supports many data sources for
which an OLE DB or .Net data provider is available.
 MCT USE ONLY. STUDENT USE PROHIBITED
9-4 Performing Predictive Analysis with Data Mining

For a list of data sources supported by Analysis Services for use in data mining, see Supported Data
Sources (SSAS - Multidimensional) in Microsoft Docs:
Supported Data Sources (SSAS - Multidimensional)
http://aka.ms/Yy2cps

Data Mining Structure

The data mining structure is the central component of a data mining solution. It contains definitions for
the data from which mining models in the solution are built, including:

 The DSV which defines the source data.

 The case table and the mining structure columns that, in turn, express the data and content types.

 A training set definition.

 A testing set definition.

Different portions of the case table data may be used to train and test data mining models.

For more information about mining structures, see Mining Structures (Analysis Services - Data Mining) in
Microsoft Docs:

Mining Structures (Analysis Services - Data Mining)

http://aka.ms/Otbw5s

Case Table
The case table stores aggregations from source data; the aggregations are used by the data mining
models.

Cases may be simple, in which circumstances each row in the case table corresponds to a case—for
example, if each case relates to a customer. When cases are complex, two or more case tables are
connected with a one-to-many relationship; these are referred to as nested cases, or nested tables.

When you specify the case table, you define the data type and content type for each column. The content
type defines how the values in a column relate to one another. Defining content types makes data mining
models more accurate as, when a content type is specified, the mining model is less likely to make poor
inferences about the statistical relationship between data values.

Some examples of content types include:

 Discrete. These are values that are not part of a sequence. Discrete columns contain data that has a
finite number of values with no continuum between them. Examples of discrete data include number
of children, phone number, or gender. Discrete values can be numeric or non-numeric.

 Continuous. These represent sequences of numeric data on a scale. Examples of continuous data
include temperature and weight.

 Cyclical. These represent data that is organized into limited, ordered sets that repeat. Examples of
cyclical data include numbered days of the week or numbered months of the year.

For a full list of content types, see Content Types (Data Mining) in Microsoft Docs:

Content Types (Data Mining)

http://aka.ms/Fpba6j
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 9-5

Data Mining Model

A data mining structure may be referenced by one or more data mining models. Each data mining model
uses a data mining algorithm, which you specify when you create the model. The mining model uses the
algorithm to analyze data from the data mining structure. When you create a data mining model, you
define the columns from the data mining structure, and specify a usage value for each column. Columns
from the data mining structure for which you do not specify a usage value are not included in the model.
The available usage values are:

 Key. This indicates a key column containing values that identify each row uniquely.

 Input. This indicates that the model should use this column to help forecast values for the predictable
column.
 Predict. This indicates the column for which you want to predict values in the mining mode. Its value
is also used as an input.

 Predict Only. This indicates the column for which you want to predict values in the mining mode. Its
value is not used as an input.

For example, if you want to predict the amount of money each customer is likely to spend on a
supermarket website, you could use CustomerID as the key column; CustomerSpend as the predictable
column; and various others, containing data such as address, age, and number of children, as input
columns.

For more information on data mining models, see Mining Models (Analysis Services - Data Mining) in
Microsoft Docs:

Mining Models (Analysis Services - Data Mining)

http://aka.ms/ipkyve

Data Mining Algorithms

Data mining algorithms provide rules for the
analysis of data. Analysis Services provides nine
algorithms that you can use to create data mining
models. You can also add further algorithms using
third-party plugins if necessary.

Widely-used data mining algorithms fall into the

following broad categories:

 Classification algorithms. These predict one

or more discrete variables based on other
attributes. Microsoft Decision Trees is an
example that might be used to forecast
whether a customer will purchase a particular
product. Microsoft Neural Network and Microsoft Naïve Bayes are other classification algorithms.
The best algorithm for your purposes depends on a number of factors, such as the volume of data or the
specific types of column being analyzed. For example, the Naïve Bayes algorithm can use discrete columns
(such as City) to classify data but does not support continuous columns that may be grouped into ranges
(such as Age). If you need to classify data into groups based on ranges of continuous values, the Decision
Trees algorithm may be a better choice.
 MCT USE ONLY. STUDENT USE PROHIBITED
9-6 Performing Predictive Analysis with Data Mining

 Regression algorithms. These predict one or more continuous variables, such as profit or loss. The
Microsoft Time Series algorithm is an example that might be used to determine a retail store’s
seasonal sales for the coming year. Microsoft Linear Regression and Microsoft Logistic Regression can
also predict continuous variables.
 Segmentation or clustering algorithms. These divide data into groups or clusters of items that have
similar properties. Microsoft Clustering, for example, might be used to divide customers into groups
with similar purchasing habits or preferences.

 Association algorithms. These find correlation between different attributes in a data set. The
Microsoft Association algorithm is an example that might be used to describe which items—such as
products being purchased together—are likely to appear in a transaction.

 Sequence analysis algorithms. Sequence analysis finds common sequences in data. The Microsoft
Sequence Clustering algorithm is an example that might be used to find common web clickthrough
paths, or the order of placing items in a cart.

For more information about the data mining algorithms included in Analysis Services, see Data Mining
Algorithms (Analysis Services - Data Mining) in Microsoft Docs:

Data Mining Algorithms (Analysis Services - Data Mining)

http://aka.ms/Nfg2r8

For more information about the plugin architecture for adding third-party data mining algorithms, see
Plugin Algorithms in Microsoft Docs:

Plugin Algorithms
http://aka.ms/Hndiao

Sequencing Activity
Put the following components of a data mining solution in order by numbering each to indicate the
correct sequence from source to target.

Steps

Source data

Analysis Services data source view

Data Mining Structure

Data Mining Model

 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 9-7

Lesson 2
Creating a Custom Data Mining Solution
To produce a data mining solution in Analysis Services, you first create a model that describes your
business problem. The model is then trained by running your data through algorithms that generate a
mathematical model. You can then either visually explore the mining model or create prediction queries
against it. Analysis Services data mining structures can use datasets from both relational and
multidimensional databases, and there is a set of algorithms you can use to investigate that data in
various ways.

This lesson describes how to create a data mining structure and data mining model. The lesson also
describes how to edit data models, and introduces the Data Mining Extensions (DMX) language, which
you can use to work with data mining in Analysis Services.

Lesson Objectives
After completing this lesson, you will be able to:

 Describe the tools used to create a data mining structure.

 Describe the tools used to modify a data mining structure.

 View a data mining model.

 Describe and work with DMX.

Data Mining Tools

In Visual Studio® SQL Server Data Tools (SSDT),
you can use the Data Mining Wizard to create new
data mining structures and models. You can then
use the Data Mining Designer in SSDT to
configure the structure.

Data Mining Wizard

You can use the Data Mining Wizard to define the
data mining structure and, optionally, to create
the first data mining model for the structure. The
Data Mining Wizard starts automatically when you
create a new data mining structure. When you run
the Data Mining Wizard, you can configure the
data source view, the case table, the Key, Input and Predictable columns for the training data, the column
data types and content types, and the proportion of the data to use for the testing set. If you choose to
create a data mining model at the same time as a data mining structure, you must also specify the mining
technique for the model to use.

Data Mining Designer

You can use the Data Mining Designer to configure the data mining structure and data mining model that
you created with the Data Mining Wizard. For example, you can use the Data Mining Designer to add new
data mining models, train, browse, and compare models, and to create predictions based on data mining
models.
 MCT USE ONLY. STUDENT USE PROHIBITED
9-8 Performing Predictive Analysis with Data Mining

For more information about the Data Mining Wizard, see Data Mining Wizard (Analysis Services - Data
Mining) in Microsoft Docs:
Data Mining Wizard (Analysis Services - Data Mining)
http://aka.ms/Hw9h83

Demonstration: Using the Data Mining Wizard

In this demonstration, you will see how to:

 Create a data mining project in Visual Studio.

 Create a data mining structure and a data mining model using the Data Mining Wizard.

Demonstration Steps
Create a Data Mining Project in Visual Studio

1. Ensure that the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and then
log on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd.

2. In the D:\Demofiles\Mod09 folder, run Setup.cmd as Administrator.

3. In the User Account Control dialog box, click Yes, and then wait for the script to finish.

4. Start Visual Studio 2015 and on the File menu, point to New, and then click Project.
5. In the New Project dialog box, expand Business Intelligence, then click Analysis Services, then
click Analysis Services Multidimensional and Data Mining Project.
6. In the Name box, type Mine AW then type D:\Demofiles\Mod09 in the Location box, then click
OK.

7. In Solution Explorer, right-click Data Sources, and then click New Data Source.

8. In the Data Source Wizard, on the Welcome to the Data Source Wizard page, click Next.

9. On the Select how to define the connection page, click New.

10. In the Connection Manager dialog box, in the Server name field, type MIA-SQL, in the Select or
enter a database name drop-down list, click AdventureWorksDW, and then click OK.

11. On the Select how to define the connection page, click Next.

12. On the Impersonation Information page, click Use a specific Windows user name and password,
then type the following credentials and click Next:
o User name: ADVENTUREWORKS\ServiceAcct

o Password: Pa55w.rd

13. On the Completing the Wizard page, click Finish.

14. In Solution Explorer, right-click Data Source Views, and then click New Data Source View.

15. In the Data Source View Wizard, on the Welcome to the Data Source View Wizard page, click
Next.

16. On the Select a Data Source page, ensure that Adventure Works DW is selected, and then click
Next.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 9-9

17. On the Select Tables and Views page, in the Available objects list, click ProspectiveBuyer (dbo),
hold the Ctrl key and click vTargetMail (dbo), and click the > button to move the selected objects
to the Included objects list. Then click Next.

18. On the Completing the Wizard page, in the Name field, type AW DW View, and then click Finish.

Create a Data Mining Structure and a Data Mining Model Using the Data Mining Wizard

1. In Solution Explorer, right-click Mining Structures, click New Mining Structure.

2. In the Data Mining Wizard, on the Welcome to the Data Mining Wizard page, click Next.

3. On the Select the Definition Method page, ensure that From existing relational database or data
warehouse is selected, and then click Next.

4. On the Create the Data Mining Structure page, ensure that Create mining structure with mining
model is selected, ensure that Microsoft Decision Trees is selected, and then click Next.

5. On the Select a Data Source View page, select AW DW View and click Next.

6. On the Specify Table Types page, in the vTargetMail row, select the check box in the Case column,
and then click Next.

7. On the Specify the Training Data page, in the Mining Model Structure table, select the following
columns and click Next:

o BikeBuyer: Predictable
o CustomerKey: Key

o All other columns: Input

8. On the Specify Columns’ Content and Data Type page, click Detect and review the content type
and data types found. Ensure that the content type of the Bike Buyer column is identified as
Discrete. Then click Next.

9. On the Create Testing Set page, note that the Percentage of data for testing value is 30 percent,
and then click Next.

10. On the Completing the Wizard page, in the Mining structure name box, type Purchase
Prediction.

11. In the Mining model name box, type Purchase Decision Tree, and then click Finish.

12. On the Build menu, click Deploy Mine AW. When deployment completes, close the Deployment
Progress – Mine AW window.

13. Leave Visual Studio open for the next demonstration.

 MCT USE ONLY. STUDENT USE PROHIBITED
9-10 Performing Predictive Analysis with Data Mining

Modifying Data Mining Structures and Models

There are two ways to make changes to data
mining structures and models.

Data Mining Designer

You can use the Data Mining Designer in Visual
Studio to edit data mining structures, add and edit
data mining models, test the accuracy of data
models, and view model data.

To add models to an existing data mining

structure, use the following procedure:
1. In Visual Studio, on the Mining Models tab,
click Create a related mining model, or
right-click an existing mining model, and then click New Mining Model.
2. In the New Mining Model dialog box, enter a model name and select an algorithm for the new
model.

Data Mining Client for Excel

Note: Microsoft has not released a version of the Data Mining tools for Excel 2016. The
following is applicable to Office 2010 and Office 2013 versions.

You can also use the Data Mining Client for Excel add-in to edit data mining structures and models. When
you use the Add Model to Structure Wizard in the Data Mining Client for Excel add-in to edit an
existing structure by adding a new model, you can select and configure a data mining algorithm. You can
use the Manage Mining Structures and Models dialog box to rename, delete, clear, process, export, and
import structures and models.

For more information about using the Data Mining Designer, see Data Mining Designer in Microsoft Docs:

Data Mining Designer

http://aka.ms/Qe2sgh

For more information about using the Data Mining Client for Excel to work with data mining structures,
see Advanced Modeling (Data Mining Add-ins for Excel) in Microsoft Docs:

Advanced Modeling (Data Mining Add-ins for Excel)

http://aka.ms/Bkdytk
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 9-11

Demonstration: Using the Data Mining Designer

In this demonstration, you will see how to modify a data mining structure with SSDT, by adding a new
data model.

Demonstration Steps
1. Ensure you have completed the previous demonstrations in this module.

2. In Visual Studio Solution Explorer, expand Mining Structures and double-click Purchase
Prediction.dmm.

3. In the Purchase Prediction.dmm [Design] window, click Mining Models.

4. Right-click anywhere in the Mining Models tab, then click New Mining Model.

5. In the New Mining Model dialog box, type Purchase Naive Bayes in the Model name box.
6. In the Algorithm name box, select Microsoft Naive Bayes, then click OK.

7. In the Microsoft Visual Studio dialog box, read the information message about content types
unsupported by the Microsoft Naïve Bayes algorithm, then click Yes.

8. In the Mining Models tab, in the Purchase Naive Bayes column, in the Name Style row, select
Ignore to exclude the column from the model.

9. On the Build menu, click Deploy Mine AW. When deployment completes, close the Deployment
Progress – Mine AW window.

10. Leave Visual Studio open for the next demonstration.

Viewing a Data Mining Model

From a design perspective, the results of a data
mining model can be viewed in two ways.

Mining Model Viewer

You can view the results of data mining analyses in
Visual Studio or SQL Server Management Studio
(SSMS) by using the Mining Model Viewer. The
viewing options available vary depending on the
type of model you want to view; for example, you
can use the Microsoft Tree Viewer to browse a
decision tree model. The Microsoft Tree Viewer
includes the Decision Tree and Dependency
Network tabs so you can view results in different
graphical formats.

In the Mining Model Viewer, you can refresh data model views and, depending on the viewer used,
perform other tasks, such as filter items, or adjust thresholds.

Browse Models by Using the Data Mining Client for Excel Add-In
You can also browse mining model results by using the Data Mining Client for Excel add-in. To browse a
model, click the Browse in the Model Usage area of the ribbon and then specify the model you want to
view. As with the Mining Model Viewer, the way results are displayed varies depending on the type of
model.
 MCT USE ONLY. STUDENT USE PROHIBITED
9-12 Performing Predictive Analysis with Data Mining

For more information about browsing data mining models using the Data Mining Client for Excel, see
Browsing Models in Excel (SQL Server Data Mining Add-ins) in Microsoft Docs:
Browsing Models in Excel (SQL Server Data Mining Add-ins)
http://aka.ms/I2cclr

Demonstration: Using the Data Mining Model Viewer

In this demonstration, you will see how to view the results of a data mining model in SSMS.

Demonstration Steps
1. Ensure you have completed the previous demonstrations in this module.

2. Start SQL Server Management Studio.

3. In the Connect to Server dialog box, select Analysis Services in the Server Type box, then type
MIA-SQL in the Server name box, then click Connect.
4. In Object Explorer, expand Databases, expand Mine AW, expand Mining Structures, right-click
Purchase Prediction, then click Browse.

5. In the Mining Model drop-down list, ensure that Purchase Naive Bayes is selected, and on the
Dependency Network tab, move the slider gradually from All Links to Strongest Links to see which
factors are the strongest predictors that a customer will purchase a bike.

6. On the Attribute Profiles tab, view the color-coded indicators of the values for each column when
compared to customers with a Bike Buyer value of 1 or 0.
7. On the Attribute Characteristics tab, in the Attribute drop-down list, ensure that Bike Buyer is
selected, and in the Value drop-down list, select 1. Then view the probability for each other column
value when the Bike Buyer value is 1.

8. On the Attribute Discrimination tab, in the Attribute drop-down list, ensure that Bike Buyer is
selected. In the Value 1 drop-down list, select 1, and in the Value 2 drop-down list, select 0. Then
note how values for all other columns favor a particular Bike Buyer value.
9. In the Mining Model drop-down list, select Purchase Decision Tree, and on the Decision Tree tab,
in the Background drop-down list, select 1. Then view the decision tree to see how the other column
values influence a value of 1 for Bike Buyer.

10. Close the Purchase Prediction [Browse] window, and leave SQL Server Management Studio open for
the next demonstration.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 9-13

What Is DMX?
Analysis Services includes the DMX query
language, which you can use to work with data
mining models. DMX is an extension of Transact-
SQL, and can be used to create the structure of
new data mining models, to train these models,
and to browse, manage, and use them to generate
predictions.

You can use the result of any DMX query as the

basis of a report. By taking advantage of the
parameterization and formatting features available
in Reporting Services, you can deliver data mining
results in a format that is easy to consume.

DMX is composed of Data Definition Language (DDL) statements, Data Manipulation Language (DML)
statements, and a set of functions and operators.

DDL Statements
You use DDL statements to create and delete data mining structures and models, and to import and
export data mining structures and models. DDL statements include CREATE, ALTER, DROP, IMPORT,
EXPORT, and SELECT INTO statements.
The following code example creates a new mining model that uses the Microsoft Naïve Bayes algorithm.
The Bike Buyer column is defined as the predictable attribute.

DMX DDL
CREATE MINING MODEL [NBSample]
(
CustomerKey LONG KEY,
Gender TEXT DISCRETE,
[Number Cars Owned] LONG DISCRETE,
[Bike Buyer] LONG DISCRETE PREDICT
)
USING Microsoft_Naive_Bayes

DML Statements
You use DML statements to train and browse mining models, and create predictions against them. DML
statements include SELECT statements, SELECT statements with PREDICTION JOIN clauses, and INSERT
INTO statements for training data mining models.
The following code example returns the midpoint, minimum age, and maximum age for all the values in
the Age column:

DMX DML
SELECT DISTINCT [Age] AS [Midpoint Age],
RangeMin([Age]) AS [Minimum Age],
RangeMax([Age]) AS [Maximum Age]
FROM [TM Decision Tree]

To help you build DMX prediction queries, SSMS includes a query builder tool and DMX templates you
can use as the basis for new queries. In the SQL Server Data Tools environment, you access the query
builder tool from the Mining Model Prediction tab of Data Mining Designer.
 MCT USE ONLY. STUDENT USE PROHIBITED
9-14 Performing Predictive Analysis with Data Mining

For more information about DMX, see Data Mining Extensions (DMX) Reference in Microsoft Docs:

Data Mining Extensions (DMX) Reference

http://aka.ms/Kapx7e

Demonstration: Querying a Mining Model Using DMX

In this demonstration, you will see how to query a data mining model using DMX and the query builder
tool.

Demonstration Steps
1. In SSMS Object Explorer, right-click the Purchase Prediction mining structure and click Build
Prediction Query.

2. In the query designer, in the Mining Model pane, click Select Model.

3. In the Select Mining Model dialog box, expand Purchase Prediction, click Purchase Naive Bayes,
and click OK.
4. In the Select Input Table(s) pane, click Select Case Table.

5. In the Select Table dialog box, click ProspectiveBuyer (dbo), and click OK.
6. Under the Mining Model pane, in the Source column, select ProspectiveBuyer table, and then in
the Field column, select EmailAddress.

7. Under the row you just added, in the Source column, click Purchase Naive Bayes mining model, in
the Field column, click Bike Buyer, and in the Criteria/Argument column, type =1.
8. Under the row you just added, in the Source column, click Prediction Function, in the Field column,
click PredictProbability, in the Alias column, type Purchase Probability.
9. In the Mining Model pane, in the Purchase Naïve Bayes model, drag the Bike Buyer column to the
Criteria/Argument column so it contains the value [Purchase Naive Bayes].[Bike Buyer].

10. On the Mining Model menu, click Query to view the DMX code that has been generated.
11. On the Mining Model menu, click Result to view the query results. The query returns the email
address of every prospective customer who is predicted to buy a bike, along with the probability
(expressed as a percentage in fraction format) that this forecast is accurate.

12. Close the Purchase Prediction [Query] window, and leave SQL Server Management Studio open for
the next demonstration.

Verify the correctness of the statement by placing a mark in the column to the right.

Statement Answer

True or false? All the data

mining algorithms
implemented by default in
Analysis Services can process
all content types.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 9-15

Lesson 3
Validating a Data Mining Model
Validation is the process of assessing how well your mining models perform against real data. It is
important to validate your mining models by understanding their quality and characteristics before you
deploy them into a production environment.

This lesson provides an overview of data mining model validation and describes the criteria for validating
models. It also introduces the tools for model validation that are provided in SQL Server Data Tools and
the Data Mining Client for Excel add-in.

Lesson Objectives
After completing this lesson, you will be able to:

 Explain the need for model validation.

 Use the available accuracy charts to validate data mining models.

Overview of Data Mining Validation

Data mining algorithms use statistical models to
predict the outcome of defined scenarios.
However, if you use multiple algorithms to analyze
a single data set, the results will not be identical.
This is because each algorithm uses different
techniques to analyze data. Consequently, when
you analyze data using different algorithms, you
need to test the results to discover which
algorithm is the most accurate for the given
scenario. This process is known as validation.
When you create a data model, you use a set of
data called the training set to build the model.
You also define a proportion of the data to use to test the model—this is known as the testing set. To test
the validity of a data mining model, you create prediction queries to use against the testing set, and then
use validation tools to compare the results of the mining model to this data. You can view the results of
validation tests in accuracy charts, which compare predicted values with actual values.

Other factors can affect the validity of a data model, including the quality of the data and the usefulness
of the results to the business.

Accuracy
Data mining models correlate a defined outcome, such as the amount of money a customer will spend,
with input criteria, such as customer age, location, number of items purchased, and so on. The accuracy of
these correlations can be affected by missing, approximate, or incorrect data. It is important to assess the
accuracy of a model so you know how much you can trust the results produced. If you are aware in
advance that a certain percentage of the data contains inaccuracies, you can decide to accept this degree
of inaccuracy in the models you create. Data that qualifies as highly accurate is not necessarily reliable or
useful. For example, you might test a model that predicts future sales for a store, based on previous
performance. The model might be very accurate because the correlations are strong. However, if the
method of calculating profitability used in the model is incorrect, the data produced is neither reliable nor
useful.
 MCT USE ONLY. STUDENT USE PROHIBITED
9-16 Performing Predictive Analysis with Data Mining

Reliability
To be regarded as reliable, a data mining model must perform consistently. To assess reliability, you need
to test the data model by using multiple, similar data sets. For each data set, the model should produce
similar results. For example, you could test a model that predicts future sales performance for a store by
using it against data from multiple stores. If one store calculates profitability differently, this will show up
in the testing results.

Usefulness
A data model might be both accurate and reliable, but this does not necessarily mean that the results it
produces are useful to the business. For example, your model might identify a strong correlation between
the profitability of a store and its location, but this might not be useful because it doesn’t explain why that
correlation exists.
For more information about validation of data mining models, see Testing and Validation (Data Mining) in
Microsoft Docs:

Testing and Validation (Data Mining)

http://aka.ms/I516n8

Accuracy Charts
Visual Studio, the Data Mining Add-In for Excel,
and SQL Server Management Studio include tools
for validating data mining models.

Visual Studio
You can use the Mining Accuracy Chart tab in the
Data Mining Designer, in the SSDT, to create
charts and reports to help you assess the validity
of data models.

 Lift chart. This is a graphical representation of

the difference between the results produced
by a data mining model and those that
random guessing would generate. The difference between these two sets of results is called the lift.
You can compare multiple models simultaneously by using a lift chart, which means you can easily
identify the best model. For example, if you are creating a mailing campaign, you can create a lift
chart to identify the model that best predicts the customers most likely to respond. Typically, a lift
chart is at its best when used with continuous data values.
 Scatter plot. This is a graphical representation of the accuracy of model predictions. Each point on
the graph represents a prediction; predicted values are displayed along the Y axis, actual values are
displayed along the X axis. The accuracy of each prediction can be determined from its proximity to
the ideal prediction line, which can be visualized as a line running at 45 degrees between the X axis
and the Y axis. For example, in a mailing campaign, in a model predicting sales by day, you could use
a scatter plot to compare the predicted and actual values of sales. A scatter plot can only be used
with continuous numeric values; it cannot be used with time series data.

 Profit chart. This is a graph that displays the estimated profit or loss identified by the model for a
particular scenario. When you create a profit chart, you specify values for the number of rows to use
in the assessment, the overall cost of the scenario, the cost per row, and the anticipated revenue per
row. For example, in a mailing campaign, you could specify to use 5,000 customer rows, the total cost
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 9-17

of the campaign, the cost per customer, and the revenue per customer. These values are used to
calculate the campaign’s profitability.
 Classification matrix. This uses testing data to count and display the number of true and false
positives that a data mining model identifies for a given predicted value. For example, imagine a
Customers table that includes a BikeBuyer column containing values of 1 and 0. The value 1
indicates that the customer purchased a bike, and 0 that they did not. A classification matrix will use
the input columns to calculate whether a customer will be a bike buyer or not, and then compare
these values to the real values in the data. The matrix will display the results, showing the number of
correct and incorrect predictions for each value. Typically, a classification matrix is more accurate
when used with discrete data values.

 Cross-validation report. Lift charts, profit charts, and classification matrices are all types of chart that
assess and display the accuracy of models in different ways. A cross-validation report is a different
way of validating a data mining model. To perform cross-validation, Analysis Services first divides a
data set into partitions. It then uses a model to analyze one of the partitions, treating this as the
training set. The other partitions are used to validate the results of this analysis. The process is then
repeated several times, using a different partition as the training set on each occasion.

SQL Server Management Studio

You can use SQL Server Management Studio to connect to an Analysis Services database containing a
data mining structure, and perform the following validation tasks for models:

 Create a lift chart.

 Create a scatter plot.

 Create a profit chart.

 Create a classification matrix.

 Create a cross-validation report.

Data Mining Client for Excel

You can use the Data Mining Client for Excel to validate data mining models using a set of wizards, which
you can access from the Accuracy and Validation area of the Data Mining tab on the ribbon.

 The Accuracy Chart Wizard enables you to create a lift chart or a scatter plot.

 The Classification Matrix Wizard enables you to create a classification matrix.

 The Profit Chart Wizard enables you to generate a profit chart.

 The Cross-Validation Wizard enables you to perform cross-validation.

For more information about using the Data Mining Client for Excel to carry out data model validation, see
the Test, Query, and Validate Models section of Data Mining Client for Excel (SQL Server Data Mining Add-
ins) in Microsoft Docs:

Data Mining Client for Excel (SQL Server Data Mining Add-ins) - Test, Query, and Validate
Models
http://aka.ms/Dlgmem
 MCT USE ONLY. STUDENT USE PROHIBITED
9-18 Performing Predictive Analysis with Data Mining

Demonstration: Viewing Accuracy Charts

In this demonstration, you will see how to view:

 A Lift Chart.

 A Profit Chart.
 A Classification Matrix.

 A Cross-Validation Report.

Demonstration Steps
Create a Lift Chart

1. Ensure you have completed the previous demonstrations in this module.

2. In SSMS Object Explorer, right-click the Purchase Prediction mining structure and click View Lift
Chart.

3. On the Input Selection tab, note that both mining models are selected with the Bike Buyer column
as predictable, and that the test cases defined in the models themselves will be used for the
validation.

4. Click the Lift Chart tab, and view the lift chart, which compares accuracy for the two mining models
you have created against an ideal model by plotting the number of correct predictions against the
number of cases in the overall sample. As the number of cases increases, the ideal model maintains an
accuracy of 100 percent. However, the models you have created tend to become less accurate the
more cases there are.

5. Review the scores in the Mining Legend pane to see which of your models is the most accurate for
this test data.

Create a Profit Chart

1. In the Chart type drop-down list, select Profit Chart.

2. In the Profit Chart Settings dialog box, enter the following values to reflect a marketing campaign
you are planning, and then click OK:

o Population: 20,000 (this is the number of potential customers you plan to contact).

o Fixed cost: 1,000 (this is the fixed cost of your marketing campaign).

o Individual cost: 3 (this is the cost associated with contacting each customer).
o Revenue per individual: 10 (this is the amount you expect a customer to spend if they respond
positively to the campaign).
3. Review the chart and the Mining Legend pane to evaluate which mining model is likely to generate
the most profitable marketing campaign based on the test data.

Create a Classification Matrix

1. Click the Classification Matrix tab.

2. Review the matrix, noting that for each model it shows the number of times the model predicted a
Bike Buyer value of 1 or 0 on rows, with columns for the actual value of the Bike Buyer column in
the test data.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 9-19

Create a Cross Validation Report

1. Click the Cross Validation tab.

2. Enter the following values, and click Get Results:

o Fold Count: 5 (this is the number of partitions used to group the data for analysis).

o Max Cases: 5 (this is the number of cases to be analyzed).

o Target Attribute: Bike Buyer (this is the predictable column to be evaluated).

o Target State: 1 (this is the desired value for the target attribute).

o Target Threshold: 0.1 (this is a value between 0 and 1 that indicates the level of accuracy
required for a prediction to be considered correct).

3. View the resulting report, and note that for each mining model, the results include the following:

o Classifications for true positives, false positives, true negatives, and false negatives.

o The likely lift gained by using the model.

4. Close SQL Server Management Studio without saving any changes.

Check Your Knowledge

Question

Which data model accuracy chart type would

you use to compare the accuracy of several
data models?

Select the correct answer.

Scatter plot

Classification matrix

Lift chart

Cross-validation report

Profit chart
 MCT USE ONLY. STUDENT USE PROHIBITED
9-20 Performing Predictive Analysis with Data Mining

Lesson 4
Connecting to and Consuming a Data Mining Model
Results produced by data mining models need to be displayed in useful and comprehensible ways, for the
benefit of data and business analysts.

This lesson describes the options for viewing data mining results and how you can use Reporting Services
reports to render them in a user-friendly format.

Lesson Objectives
After completing this lesson, you will be able to:

 Use Reporting Services to display data mining results.

Creating a Connection to a Data Mining Model from a Reporting Services

Report
You have seen how to use DMX queries to
generate predictive information from a data
mining model. To provide a way of viewing data
mining results that is easier for business users to
consume, you can create a SQL Server Reporting
Services (SSRS) report that uses DMX to query the
data mining mode and provide predictive
information. SSRS reports offer a wide range of
display and formatting options that enable you to
show results in the most intuitive way. For
convenience, you can also incorporate the results
of multiple models into a single report.

Data mining models store data in an Analysis Services database. To connect to an Analysis Services
database and create a data mining report, you create an Analysis Services connection in an SSRS project,
in SSDT. You can then use Query Builder to create the query that retrieves the data to use in the report.

For more information about using DMX to query data models in an SSRS report, see Retrieve Data from a
Data Mining Model (DMX) (SSRS) in Microsoft Docs:

Retrieve Data from a Data Mining Model (DMX) (SSRS)

http://aka.ms/E2h6fz
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 9-21

Demonstration: Using Data Mining Results in an SSRS Report

In this demonstration, you will see how to include data mining results in an SSRS report.

Demonstration Steps
1. Ensure you have completed the previous demonstrations in this module.

2. In Visual Studio, on the File menu, point to New and click Project.

3. In the New Project dialog box, expand Business Intelligence, select Reporting Services, select
Report Server Project Wizard.
4. In the Name field, type Purchase Prediction, in the Location field, type D:\Demofiles\Mod09, and
then click OK.

5. In the Report Wizard, on the Welcome to the Report Wizard page, click Next.
6. On the Select the Data Source page, select New data source, change the Name to AWMine, in the
Type drop-down list, select Microsoft SQL Server Analysis Services, and click Edit.
7. In the Connection Properties dialog box, in the Server name field, type MIA-SQL, in the Select or
enter a database name drop-down list, click Mine AW, and then click OK.

8. On the Select the Data Source page, click Next.

9. On the Design the Query page, click Query Builder.

10. In Query Designer, in the Mining Model pane, click Select Model.

11. In the Select Mining Model dialog box, expand Purchase Prediction, click Purchase Naive Bayes,
and click OK.
12. In the Select Input Table(s) pane, click Select Case Table.

13. In the Select Table dialog box, click ProspectiveBuyer (dbo), and click OK.

14. Under the Mining Model pane, in the Source column, select ProspectiveBuyer table, and then in
the Field column, select EmailAddress.

15. Under the row you just added, in the Source column, click Purchase Naive Bayes mining model, in
the Field column, click Bike Buyer, and in the Criteria/Argument column, type =1.

16. Under the row you just added, in the Source column, click Prediction Function, in the Field column,
click PredictProbability, in the Alias column, type Purchase Probability.

17. in the Mining Model pane, from the Purchase Naïve Bayes model, drag the Bike Buyer column to
the Criteria/Argument column so it contains the value [Purchase Naive Bayes].[Bike Buyer], then
click OK.

18. On the Design the Query page, click Next.

19. On the Select the Report Type page, verify that Tabular is selected, then click Next.

20. On the Design the Table page, in the Available fields box, click EmailAddress, then click Details >.

21. In the Available fields box, click Purchase_Probability, then click Details >. Click Next.

22. If the Choose the Table Style page appears, click Next.

23. If the Choose the Deployment Location page appears, click Next.
24. On the Completing the Wizard page, type Likely Purchase Email Report in the Report name box,
then click Finish.
 MCT USE ONLY. STUDENT USE PROHIBITED
9-22 Performing Predictive Analysis with Data Mining

25. In the report designer, on the Design tab, click the table.

26. In the Row Groups pane, right-click the (table1_Details_Group) drop-down list and click Group
Properties.

27. In the Group Properties dialog box, on the Sorting page, click Add, in the Sort by drop-down list,
select [Purchase_Probability], and in the Order drop-down list, click Z to A. Then click OK.

28. In the Purchase Probability column, right-click [Purchase Probability] and click Text Box
Properties.

29. In the Text Box Properties dialog box, click Number, select Percentage, and click OK.
30. Click the Preview tab to review the report.

31. Close Visual Studio.

Verify the correctness of the statement by placing a mark in the column to the right.

Statement Answer

True or false? You can

include the results of
multiple mining models in a
single SSRS report.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 9-23

Lab: Using Data Mining to Support a Marketing Campaign

Scenario
The marketing department at Adventure Works Cycles is planning a direct mail campaign. To maximize its
effectiveness, you have been asked to create a report that uses data mining techniques to identify the
subset of potential customers who are most likely to purchase a bike.

Objectives
After completing this lab, you will be able to:

 Create a data mining structure and a data mining model, then add a data mining model to a data
mining structure.

 Explore a data mining model.

 Validate data mining models.

 Use a data mining model as a data source for a report.

Estimated Time: 60 minutes

Virtual machine: 20768C-MIA-SQL

User name: ADVENTURWORKS\Student

Password: Pa55w.rd

Exercise 1: Creating a Data Mining Structure and Models

Scenario
To support the marketing campaign, you want to create a data mining model that will help you to
identify the individuals most likely to buy a bike. You will use SQL Server Data Tools to create the structure
and model, and then deploy the model to Analysis Services. When this is complete, you will add a second
data model to the structure.

The main tasks for this exercise are as follows:

1. Prepare the Lab Environment

2. Create a Data Mining Project

3. Create a Data Mining Structure and a Data Mining Model

4. Add a Data Mining Model to a Data Mining Structure

 Task 1: Prepare the Lab Environment

1. Start the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines, and then log on to 20768C-
MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd.

2. Run Setup.cmd in the D:\Labfiles\Lab09\Starter folder as Administrator.

 Task 2: Create a Data Mining Project

1. Use Visual Studio to create an Analysis Services multidimensional and data mining project named AW
Data Mining. Save the project in the D:\Labfiles\Lab09\Starter folder.
2. Create a data source named Adventure Works DW that connects to the AdventureWorksDW
database on the MIA-SQL Database Engine instance. Use the Service Account.
 MCT USE ONLY. STUDENT USE PROHIBITED
9-24 Performing Predictive Analysis with Data Mining

3. Use the Adventure Works DW data source to create a data source view named Adventure Works
DW DM View that includes the ProspectiveBuyer table and the vTargetMail view.

 Task 3: Create a Data Mining Structure and a Data Mining Model

1. Use the Data Mining Wizard to create a data mining structure from existing data that uses the
Microsoft Decision Trees algorithm.

2. Specify the vTargetMail view from the Adventure Works DW DM View data view as the case table.

3. Configure the columns in the case table as follows:

o BikeBuyer: Predictable

o CustomerKey: Key
4. All other columns: Input.

5. Have the wizard detect the content and data types of the columns. Ensure that the content type of
the Bike Buyer column is identified as Discrete, and change the Yearly Income column to Discrete.

6. Create a testing set with 30 percent of the data.

7. Name the data mining structure Purchase Prediction, and name the data mining model Purchase
Decision Tree.

8. Deploy the completed model to the MIA-SQL Analysis Services instance.

 Task 4: Add a Data Mining Model to a Data Mining Structure

1. Use Visual Studio to add a new mining model to the Purchase Prediction mining structure. The
model should use the Microsoft Naive Bayes mining algorithm.

2. Call the model Purchase – Bayes, and amend the model definition to ignore the Name Style
column.

3. Deploy the completed model to the MIA-SQL Analysis Services instance.

4. Leave Visual Studio open for the next exercise.

Results: After this exercise, you should have a data mining structure with two data models.

Exercise 2: Explore a Data Mining Model

Scenario
Now that the data mining structure has been deployed, you will use the Visual Studio Mining Model
Viewer to explore and review data models.

The main tasks for this exercise are as follows:

1. Review Data Models

 Task 1: Review Data Models

1. In Visual Studio, use the Mining Model Viewer to explore the output of the Purchase Decision Tree
and Purchase – Bayes models in the Purchase Prediction mining structure. Notice that different
viewers are available for data models based on different data mining algorithms.

2. Leave Visual Studio open for the next exercise.

 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 9-25

Results: After this exercise, you should have reviewed the output of data mining models using Visual
Studio.

Exercise 3: Validating Data Mining Models

Scenario
Now you have created the data mining models, you want to validate them to see which is the most
accurate for your data. To do this, you will use Visual Studio.

The main tasks for this exercise are as follows:

1. Review a Lift Chart

2. Review a Profit Chart

3. Review a Classification Matrix

4. Review a Cross-Validation Report

 Task 1: Review a Lift Chart

 In Visual Studio, review a lift chart for the data models in the Purchase Prediction mining structure.

Which of the models shows greater accuracy?

 Task 2: Review a Profit Chart

 Review a profit chart for the data models in the Purchase Prediction mining structure. Use the
following values to configure your profit chart:

o Population: 5,000

o Fixed cost: 500

o Individual cost: 1

o Revenue per individual: 150

 Task 3: Review a Classification Matrix

 Review a classification matrix for the data models in the Purchase Prediction mining structure.

 Task 4: Review a Cross-Validation Report

1. Review a cross-validation report for the data models in the Purchase Prediction mining structure.
Use the following values to configure your cross-validation matrix:

o Fold Count: 5

o Max Cases: 500

o Target Attribute: Bike buyer

o Target State: 1

o Target Threshold: 0.7

Leave Visual Studio open for the next exercise.

Results: After this exercise, you should have two validated data mining models in your data mining
structure.
 MCT USE ONLY. STUDENT USE PROHIBITED
9-26 Performing Predictive Analysis with Data Mining

Exercise 4: Using a Data Mining Model in a Report

Scenario
Now you have validated the data mining models, you want to create a report that contains the list of
potential bike purchasers, listed from “most likely to purchase” to “least likely”. You will create a Reporting
Services report that uses data from the AW Data Mining database, and then format the report.

The main tasks for this exercise are as follows:

1. Create a Report

 Task 1: Create a Report

1. Use Visual Studio to create a new Report Server Project Wizard project named Promotion
Targeting in the D:\Labfiles\Lab09\Starter folder.

2. Create a new data source for the AW Data Mining database in the MIA-SQL instance of Analysis
Services.

3. Use the query builder to create a query that uses the Purchase - Bayes model and the
ProspectiveBuyer(dbo) case table. Configure the query to return the following fields:

Source Field Alias Criteria/Argument

ProspectiveBuyer table LastName

ProspectiveBuyer table Address Line 1

ProspectiveBuyer table City

Purchase – Bayes mining Bike Buyer =1

model

Prediction Function PredictProbability Purchase [Purchase -

Probability Bayes].[Bike
Buyer]

4. Create a tabular report with all fields in the details section, format it using any style, and name it
Potential Bike Buyers.

5. Sort the report from Z to A by the Purchase Probability column and change the format of the
Purchase Probability column to percentage.

6. Preview the Potential Bike Buyers report, and modify its formatting until you are happy with it.

7. When you have finished working on the report, close Visual Studio.

Results: After this exercise, you should have an SSRS report that predicts bike purchasers.
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 9-27

Check Your Knowledge

Question

Which type of data mining algorithm

would you use to identify sequences
in your data?

Select the correct answer.

A regression algorithm, such as the

Microsoft Time Series algorithm.

A sequence analysis algorithm, such

as the Microsoft Sequence
Clustering algorithm.

A classification algorithm, such as

the Microsoft Decision Trees
algorithm.

A clustering algorithm, such as the

Microsoft Clustering algorithm.

An association algorithm, such as

the Microsoft Association algorithm.
 MCT USE ONLY. STUDENT USE PROHIBITED
9-28 Performing Predictive Analysis with Data Mining

Module Review and Takeaways

In this module, you have learned how to use the data mining capability of SQL Server Analysis Services for
predictive analysis.

Review Question(s)
Question: How have you seen, or can you envisage, data mining being used in organizations
where you have worked?
 MCT USE ONLY. STUDENT USE PROHIBITED
Developing SQL Data Models 9-29

Course Evaluation

Your evaluation of this course will help Microsoft understand the quality of your learning experience.

Please work with your training provider to access the course evaluation form.
Microsoft will keep your answers to this survey private and confidential and will use your responses to
improve your future learning experience. Your open and honest feedback is valuable and appreciated.
MCT USE ONLY. STUDENT USE PROHIBITED
 
 MCT USE ONLY. STUDENT USE PROHIBITED
L1-1

Module 1: Introduction to Business Intelligence and Data

Modeling
Lab: Exploring a Business Intelligence
Solution
Exercise 1: Exploring a Data Warehouse
 Task 1: Prepare the Lab Environment
1. Ensure that the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and then
log on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd.

2. In the D:\Labfiles\Lab01\Starter folder, right-click Setup.cmd, and then click Run as

administrator.

3. Click Yes when prompted to confirm that you want to run the command file, and then wait for the
script to finish.

 Task 2: Explore a Data Warehouse Schema

1. Start SQL Server Management Studio and connect to the MIA-SQL instance of the database engine.
Use Windows® authentication.

2. In Object Explorer, expand Databases, and then expand AdventureWorksDW.

3. Right-click Database Diagrams, and then click New Database Diagram.

4. In the Add Table dialog box, click DimAccount, hold down Shift and click FactSurveyResponse,
click Add, and then click Close.

5. Explore the diagram, noting the following details:

o The FactInternetSales table stores details of sales orders that are made through the Adventure
Works website. The table is related to the DimCustomer table, which contains details of the
customers who have placed orders.
o The FactResellerSales table stores details of sales orders that are made to resellers. The table is
related to the DimReseller table, which contains details of the resellers who have placed orders.

o The DimDate table stores values for dates, including details of the calendar and fiscal periods in
which individual dates occur.

o The FactInternetSales and FactResellerSales tables are related to the DimDate table by
multiple fields. These fields represent the order date (the date when the order was placed), the
due date (the date when the order was expected to be in stock), and the ship date (the date
when the order was shipped to the customer).

o Both the DimCustomer and DimReseller tables are related to the DimGeography table, which
stores details of geographical locations. The DimSalesTerritory table is also related to
DimGeography.

o The DimProduct table contains details of products, and is related to the

DimProductSubcategory table, which is in turn related to the DimProductCategory table.

o Many tables include multiple language values for the same data value, for example, the DimDate
table stores the English, French, and Spanish words for each month name.
 MCT USE ONLY. STUDENT USE PROHIBITED
L1-2 Developing SQL Data Models

o The DimEmployee table is related to itself to represent the fact that each employee has a
manager, who is also an employee.
6. On the File menu, click Save Diagram_0, and then save the diagram as AdventureWorksDW
Schema.

 Task 3: Query a Data Warehouse

1. Click File, click Open, click File, browse to the D:\Labfiles\Lab01\Starter folder, and then double-
click Query DW.sql.

2. Select the Transact-SQL code under the comment Internet sales by year and month, and then click
Execute.

3. View the results of the query, and note the following details:

o The data warehouse contains historical Internet sales orders from July 2005 to July 2008. Reseller
sales in the data warehouse are also recorded for this time period.
o You can use the MonthNumberOfYear column in the DimDate table to sort month names into
chronological order—without this field, it would be difficult (though not impossible) for reporting
clients to sort months other than alphabetically. A similar field named DayNumberOfWeek can
be used to sort weekday names into chronological order.

4. Select the Transact-SQL code under the comment Geographical reseller sales, and then click
Execute.
5. View the results of the query, and note the following details:

o In 2005, Adventure Works only sold to resellers in the United States and Canada.

o In 2006, this was expanded to include France and the United Kingdom.
o In 2007, resellers in Australia and Germany were added.

o By contrast, Adventure Works has sold directly to Internet customers in all of these regions since
2005.
6. Select the Transact-SQL code under the comment Sales by product category, and then click
Execute.

7. View the results of the query, and note the following details:

o Adventure Works sells four categories of product: Accessories, Bikes, Clothing, and Components.

o Components are only sold to resellers, not to Internet customers.

o Accessories were not sold to Internet customers until 2007.

Results: After completing this exercise, you will have:

Used a database diagram to explore a data warehouse schema.

Used Transact-SQL queries to explore the data in a data warehouse.

 MCT USE ONLY. STUDENT USE PROHIBITED
L1-3

Exercise 2: Exploring a Data Model

 Task 1: View a SQL Server Analysis Services Database
1. In SQL Server Management Studio, in Object Explorer, click Connect, and then click Analysis
Services.

2. In the Connect to Server dialog box, in the Server name field, type MIA-SQL, and then click
Connect.

3. In Object Explorer, under the MIA-SQL Microsoft Analysis Server node, expand Databases, and then
expand Adventure Works OLAP.

4. Under the Adventure Works OLAP database, expand Data Sources. Note that the SQL Server
Analysis Services database includes a data source called Adventure Works Data Warehouse, which
connects to the AdventureWorksDW database that you explored in the previous exercise.

5. Under the Adventure Works OLAP database, expand Cubes. Right-click Internet Sales, and then
click Browse.

6. In the cube browser window, in the Metadata pane, expand Measures, expand Internet Sales
Order, and then drag the Sales Amount measure onto the empty query pane. Click on the text Click
to execute the query, the total Internet sales amount is displayed.
7. In the Metadata pane, expand Order Date, and then drag Order Date.Fiscal Year to the query
pane. Click on the text Click to execute the query, the sales revenue is now aggregated by fiscal
year. In Adventure Works, fiscal years run from July to June, so fiscal year 2006 represents the period
from July 2005 to June 2006.

8. Close SQL Server Management Studio without saving any files.

 Task 2: Create a PivotTable in Excel

1. Start Excel 2016, and then create a new blank workbook.
2. On the Data tab, in the Get External Data area or list, in the From Other Sources list, click From
Analysis Services.

3. In the Data Connection Wizard dialog box, in the Server name box, type MIA-
SQL.adventureworks.msft, ensure that Use the following User Name and Password is selected,
type the following, and then click Next:

a. User Name: ADVENTUREWORKS\Student

b. Password: Pa55w.rd
4. On the Select Database and Table page, in the Select the database that contains the data you
want drop-down list, select the Adventure Works OLAP database, in the Connect to a specific
cube or table field, click Internet Sales, and then click Next.

5. On the Save Data Connection File and Finish page, click Finish.

6. In the Import Data dialog box, select PivotTable Report, ensure that Existing worksheet is selected
with the cell reference =$A$1, and then click OK.

7. In the PivotTable Fields pane, under Internet Sales Order, select Sales Amount. The total revenue
from Internet sales is displayed in the PivotTable on the worksheet.

8. In the PivotTable Fields pane, in the Values area, in the drop-down list for the Sales Amount field,
click Value Field Settings.

9. In the Value Field Settings dialog box, click Number Format.

 MCT USE ONLY. STUDENT USE PROHIBITED
L1-4 Developing SQL Data Models

10. In the Format Cells dialog box, select Accounting, click OK.

11. In the Value Field Settings dialog box, click OK. The Internet sales total in the PivotTable on the
worksheet is formatted as a currency value.

12. In the PivotTable Fields pane, under Order Date, select Order Date.Calendar Date. The total
calendar years for the sales are displayed as columns.

13. In the PivotTable Fields pane, drag Order Date.Calendar Date from the Columns area to the Rows
area. The years are now displayed on rows.
14. In the PivotTable, expand 2006. The calendar semesters for the year are shown. Expand H1CY2006 to
reveal the first two quarters, expand Q1 CY 2006 to reveal the first three months, and then expand
January to reveal daily sales totals in January 2006.

15. Save the workbook as Sales.xlsx in the D:\Labfiles\Lab01\Starter folder.

 Task 3: Filter the PivotTable

1. In Excel, in the Sales.xlsx workbook, click any cell in the PivotTable on the Sheet1 worksheet.

2. On the ribbon, on the Analyze tab, click Insert Slicer.

3. In the Insert Slicers dialog box, under Product, in the Product Category hierarchy, select English
Product Category Name, and then click OK.

4. Move the slicer so that you can see the PivotTable. Note that the Components category is disabled
because there are no sales of components to Internet customers.
5. In the slicer, click Accessories. The PivotTable is filtered to show only sales of accessories. Note that
no accessories were sold in 2005 or 2006.

6. In the slicer, click the Clear Filter icon to remove the filter.

7. Save the workbook, and then close Excel.

Results: After completing this exercise, you will have used Excel to explore an analytical data model built
on the data warehouse.
 MCT USE ONLY. STUDENT USE PROHIBITED
L2-1

Module 2: Creating Multidimensional Databases

Lab: Creating a Multidimensional Database
Exercise 1: Creating a Data Source
 Task 1: Prepare the Lab Environment
1. Ensure that the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and then
log on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd.

2. In the D:\Labfiles\Lab02\Starter folder, right-click Setup.cmd, and then click Run as

administrator.

3. Click Yes when prompted to confirm that you want to run the command file, and then wait for the
script to finish. Note that the script may take up to several minutes to complete.

 Task 2: Create an Analysis Services Project

1. On the taskbar, click Visual Studio 2015.
2. In Visual Studio, under Start, click New Project, in the New Project dialog box, in the Templates list,
click Business Intelligence, in the main window, click Analysis Services Multidimensional and
Data Mining Project, and then in the Name box, type Adventure Works OLAP.

3. Click Browse, browse to the D:\Labfiles\Lab02\Starter folder, click Select Folder, and then click OK.

 Task 3: Create a Data Source

1. In Solution Explorer, right-click the Data Sources folder, and then click New Data Source.

2. In the Data Source Wizard, on the Welcome to the Data Source Wizard page, click Next.

3. On the Select how to define the connection page, click New.

4. In the Connection Manager dialog box, in the Server name box, type localhost.

5. In the Log on to the server area, ensure that Use Windows Authentication is selected.
6. In the Connect to a database area, in the Select or enter a database name box, click
AdventureWorksDW, and then click OK.

7. On the Select how to define the connection page, click Next.

8. On the Impersonation Information page, select Use a specific Windows user name and
password.

9. In the User name box, type ADVENTUREWORKS\ServiceAcct.

10. In the Password box, type Pa55w.rd, and then click Next.

11. On the Completing the Wizard page, in the Data source name field, change the value to
Adventure Works Data Warehouse, and then click Finish.

Results: After completing this exercise, you will have:

Created an Analysis Services project by using SQL Server Data Tools.

Created a data source.
 MCT USE ONLY. STUDENT USE PROHIBITED
L2-2 Developing SQL Data Models

Exercise 2: Creating and Configuring a Data Source View

 Task 1: Create a Data Source View
1. In Solution Explorer, right-click the Data Source Views folder, and then click New Data Source
View.

2. On the Welcome to the Data Source View Wizard page, click Next.

3. On the Select a Data Source page, verify that the Adventure Works Data Warehouse data source
is selected, and then click Next.

4. In the Available objects list, click DimCustomer (dbo), and then hold down the Ctrl key and click
DimDate (dbo), DimGeography (dbo), DimProduct (dbo), DimProductCategory (dbo),
DimProductSubcategory (dbo), and FactInternetSales (dbo). Click the right arrow (>) button to
add the selected tables to the Included objects list, and then click Next.

5. On the Completing the Wizard page, in the Name field, type Adventure Works DSV, and then
click Finish. The Data Source View Designer opens automatically.

 Task 2: Configure a Data Source View

1. In the diagram, click the title bar of the FactInternetSales table and press F4.
2. In the Properties pane, change the FriendlyName property to Internet Sales, and then press Enter.

3. Click the DimCustomer table, change its FriendlyName property to Customer, and then press Enter.

4. Repeat the previous step to edit the FriendlyName property of the following tables (Table name:
FriendlyName):

o DimDate: Date

o DimGeography: Geography
o DimProduct: Product

o DimProductCategory: Product Category

o DimProductSubcategory:

o Product Subcategory

5. In the diagram, right-click the Customer table, and then click New Named Calculation.

6. In the Create Named Calculation dialog box, in the Column name box, type Full Name.

7. In the Expression box, type the text in the following code example:

CASE
WHEN MiddleName IS NULL THEN
FirstName + ' ' + LastName
ELSE
FirstName + ' ' + MiddleName + ' ' + LastName
END

8. In the Create Named Calculation dialog box, click OK.

9. On the File menu, click Save All.

Results: After completing this exercise, you will have:

Created a data source view.

Configured a data source view.

 MCT USE ONLY. STUDENT USE PROHIBITED
L2-3

Exercise 3: Creating and Configuring a Cube

 Task 1: Create a Cube
1. In Solution Explorer, right-click the Cubes folder, and then click New Cube.

2. On the Welcome to the Cube Wizard page, click Next.

3. On the Select Creation Method page, verify that Use existing tables is selected, and then click
Next.

4. On the Select Measure Group Tables page, click Suggest.

5. Note that the wizard selects the Internet Sales table as the measure group table, and then click
Next.

6. On the Select Measures page, clear every check box except Order Quantity, Total Product Cost,
Sales Amount, and Internet Sales Count, and then click Next.

7. On the Select New Dimensions page, clear the Internet Sales check box, and then click Next.

8. On the Completing the Wizard page, change the Cube name to Sales, and then click Finish. The
Cube Designer opens automatically.

 Task 2: Configure Measures

1. In Solution Explorer, click Sales.cube. In the Measures pane of the Cube Designer, expand the
Internet Sales measure group, right-click the Order Quantity measure, and then click Rename.

2. Rename Order Quantity to Internet Order Quantity, and then press Enter.
3. In the Measures pane, right-click the Total Product Cost measure, and then click Rename.

4. Rename Total Product Cost to Internet Cost, and then press Enter.

5. In the Measures pane, right-click Sales Amount, and then click Rename.

6. Rename Sales Amount to Internet Revenue, and then press Enter.

7. On the File menu, click Save All.

 Task 3: Configure Dimensions

1. In Solution Explorer, under Dimensions, right-click Customer.dim, and then click View Designer.
2. In the Data Source View pane, in the Geography table, click City.

3. Hold down the Ctrl key, click StateProvinceName, EnglishCountryRegionName, and PostalCode,
and then drag the selected columns to the Attributes pane.

4. In the Data Source View pane, in the Customer table, click CustomerAlternateKey.

5. Hold down the Ctrl key, click Title, FirstName, MiddleName, LastName, and Full Name.

6. Drag the selected columns to the Attributes pane.

7. On the File menu, click Save All then close the Customer.dim Dimension Designer.

8. In Solution Explorer, under Dimensions, right-click Product.dim, and then click View Designer.

9. In the Data Source View pane, in the Product table, click ProductAlternateKey. Hold down the Ctrl
key, click EnglishProductName and ListPrice, and then drag the selected columns to the Attributes
pane.

10. In the Data Source View pane, in the Product Subcategory table, click
EnglishProductSubcategoryName, and drag the selected column to the Attributes pane.
 MCT USE ONLY. STUDENT USE PROHIBITED
L2-4 Developing SQL Data Models

11. In the Data Source View pane, in the Product Category table, click EnglishProductCategoryName,
and drag the selected column to the Attributes pane.

12. On the File menu, click Save All then close the Product.dim Dimension Designer.

13. In Solution Explorer, under Dimensions, right-click Date.dim, and then click View Designer.
14. In the Data Source View pane, in the Date table, click FullDateAlternateKey. Hold down the Ctrl key,
click EnglishMonthName, MonthNumberOfYear, CalendarQuarter, CalendarYear, and
CalendarSemester and drag the selected columns to the Attributes pane.
15. In the Attributes pane, right-click English Month Name and click Rename. Change the attribute
name to Month.

16. On the File menu, click Save All, and then close the Date.dim designer window.

 Task 4: Browse the Cube

1. In Solution Explorer, right-click the Adventure Works OLAP solution, and then click Deploy. If an
Account Password dialog box appears, in the Password field type Pa55w.rd, and then click OK.
Wait for the Deploy Completed Successfully message in the Deployment Progress – Adventure
Works OLAP dialog box, and then close the Deployment Progress – Adventure Works OLAP
dialog box.

2. In the Sales.cube Cube Designer, click the Browser tab.

Tip: Click the Auto Hide icon on the various Visual Studio panes to make it easier to see the entire
Cube Browser window.

3. In the Measure Group pane, expand Measures, expand Internet Sales, and then drag the Internet
Revenue measure to the Drag levels or measures here to add to the query area.
4. In the Measure Group pane, drag Internet Sales Count to the right of the Internet Revenue
column.
5. In the Measure Group pane, expand the Order Date dimension, and drag the Order Date.Calendar
Year attribute to the left of the Internet Revenue column. The Cube Browser shows historical sales
amounts and counts for multiple years.

6. In Visual Studio, on the File menu, click Save All.

7. Keep Visual Studio open for the next exercise.

Results: After completing this exercise, you will have:

Created a cube.

Edited measures in the cube.

Edited dimensions in the cube.

Browsed the cube.

 MCT USE ONLY. STUDENT USE PROHIBITED
L2-5

Exercise 4: Adding a Dimension to the Cube

 Task 1: Edit the Data Source View
1. In Visual Studio, in the Adventure Works OLAP project, in Solution Explorer, under Data Source
Views, double-click Adventure Works DSV.dsv to open it. In the Data Source View menu, click
Add/Remove Tables.

2. In the Add/Remove Tables dialog box, in the Available objects list, click DimSalesTerritory, click
the right arrow (>) button, and then click OK.

3. In the diagram, click the title bar of the DimSalesTerritory table and press F4.

4. In the Properties pane, change the FriendlyName property to Sales Territory, and then press Enter.

5. On the File menu, click Save All.

 Task 2: Create a Dimension

1. In Solution Explorer, right-click the Dimensions folder and click New Dimension.
2. In the Dimension Wizard, on the Welcome to the Dimension Wizard page, click Next.

3. On the Select Creation Method page, select Use an existing table, and click Next.
4. On the Specify Source Information page, ensure that the Adventure Works DSV data source view
is selected. In the Main table list, select Sales Territory. Verify that SalesTerritoryKey is selected as
the key column and the name column, and then click Next.
5. On the Select Dimension Attributes page, select SalesTerritoryRegion, SalesTerritoryCountry,
and SalesTerritoryGroup, ensure that the Enable Browsing check box is selected for each of them,
and then click Next.
6. On the Completing the Wizard page, click Finish. The new dimension is opened in the Dimension
Designer.

 Task 3: Add a Dimension to a Cube

1. In Solution Explorer, under Cubes, double-click Sales.cube to open it in the Cube Designer, and then
click the Cube Structure tab.

2. On the Cube menu, click Add Cube Dimension.

3. In the Add Cube Dimension dialog box, click Sales Territory, and then click OK.

4. Click the Dimension Usage tab, verify that the value Sales Territory appears in the Dimensions list,
and then next to it, in the Measure Groups list, the value Sales Territory Key appears. This confirms
that the Sales Territory dimension is related to the Internet Sales measure group using the Sales
Territory Key attribute.

5. On the File menu, click Save All.

 Task 4: Analyze a Cube by Using Excel

1. In Solution Explorer, right-click the Adventure Works OLAP solution, and then click Deploy. If an
Account Password dialog box appears, in the Password field type Pa55w.rd and click OK. Wait for
the Deploy Completed Successfully message in the Deployment Progress – Adventure Works
OLAP dialog box, and then close the Deployment Progress – Adventure Works OLAP dialog box.

2. When deployment has completed successfully, in the Sales.cube Cube Designer, on the Browser tab,
click Reconnect.

3. On the Cube menu, click Analyze in Excel. If you are prompted to enable data connections, click
Enable.
 MCT USE ONLY. STUDENT USE PROHIBITED
L2-6 Developing SQL Data Models

4. In the PivotTable Fields pane, beneath the Internet Sales measure, select Internet Revenue.

5. In the PivotTable Fields pane, under the Sales Territory dimension, select Sales Territory Group.
Sales for each territory group are shown.

6. In the PivotTable Fields pane, under the Sales Territory dimension, select Sales Territory Country.
Sales within each sales territory group are broken down for each country.

7. In the PivotTable Fields pane, under the Sales Territory dimension, select Sales Territory Region.
Sales within each sales territory are broken down for each region. Note that some countries do not
include sales regions, so the country is shown at both the country and region levels.

8. Close Excel without saving the workbook.

9. In Visual Studio, on the File menu, click Save All, and then close Visual Studio.

Results: After completing this exercise, you will have:

Edited the data source view.

Created a dimension.

Added a dimension to the cube.

Browsed the cube by using Excel.

 MCT USE ONLY. STUDENT USE PROHIBITED
L3-1

Module 3: Working with Cubes and Dimensions

Lab: Working with Cubes and Dimensions
Exercise 1: Configuring Dimensions
 Task 1: Prepare the Lab Environment
1. Ensure the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and then log
on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd.

2. In the D:\Labfiles\Lab03\Starter folder, right-click Setup.cmd and click Run as administrator.

3. In the User Account Control dialog, click Yes, and wait for the script to finish.

 Task 2: Remove Unused Attributes

1. Start Visual Studio and open Adventure Works OLAP.sln in the D:\Labfiles\Lab03\Starter folder.
2. In Solution Explorer, expand Dimensions and double-click the Customer.dim dimension. Notice that
many attributes have been added to allow business users to aggregate measures in many different
ways. However, users have complained that some of these attributes are unnecessary and that they
should be removed to make browsing the cube simpler.

3. In the Attributes pane, click Commute Distance, press the Ctrl key, and click Number Cars Owned
and Number Children At Home, right-click any of the highlighted attributes, and then click Delete.
In the Delete Objects dialog box, click OK.

4. On the File menu, click Save All, then close the Customer.dim Dimension Designer.

5. In Solution Explorer, double-click the Product.dim dimension. Again, users have requested that you
remove some unnecessary attributes from this dimension.

6. In the Attributes pane, click Days To Manufacture, press the Ctrl key, and click Safety Stock Level.
Press Delete and, in the Delete Objects dialog box, click OK.

7. On the File menu, click Save All, then close the Product.dim Dimension Designer.

 Task 3: Add Dimension Intelligence

1. In Solution Explorer, right-click Date.dim, and then click Add Business Intelligence.

2. On the Welcome to the Business Intelligence Wizard page, click Next.

3. On the Choose Enhancement page, click Define dimension intelligence, and then click Next.

4. On the Define Dimension Intelligence page, in the Dimension type field, click Time.
5. In the Dimension attributes table, select the Include check box for the following attribute types,
and select the corresponding item in the Dimension Attribute column:

o Year: Calendar Year

o Half Year: Calendar Semester

o Quarter: Calendar Quarter

o Month: Full Date Alternate Key

o Date: Month

6. Click next, and then click Finish.

 MCT USE ONLY. STUDENT USE PROHIBITED
L3-2 Developing SQL Data Models

7. Leave Visual Studio open for the next lab.

Results: After this exercise, unused attributes in the Customer and Product dimensions will have been
removed; time intelligence will have been added to the Date dimension.

Exercise 2: Defining Relationships and Hierarchies

 Task 1: Create a Natural Hierarchy
1. In Solution Explorer, double-click the Product.dim dimension to open it in the Dimension Designer.
Note that this dimension includes attributes from three related tables (Product, Product
Subcategory, and Product Category).

2. On the Dimension Structure tab, in the Attributes pane, drag English Product Category Name into
an empty area of the Hierarchies pane. This creates a new hierarchy named Hierarchy.
3. In the Attributes pane, drag English Product Subcategory Name to the <new level> area beneath
English Product Category Name in the hierarchy.

4. In the Attributes pane, drag English Product Name to the <new level> area beneath English
Product Subcategory Name in the hierarchy.

Note: if a warning is displayed, notifying you that attribute relationships do not exist and
performance may be decreased, ignore it. You will see how to use attribute relationships to optimize
a hierarchy later in this lab.

5. Right-click Hierarchy and click Rename. Then rename the hierarchy to Categorized Products.

6. In the Categorized Products hierarchy, right-click English Product Category Name and click
Rename. Rename the hierarchy level to Category.

7. In the Categorized Products hierarchy, right-click English Product Subcategory Name and click
Rename then rename the hierarchy level to Subcategory.
8. In the Categorized Products hierarchy, right-click English Product Name and click Rename then
rename the hierarchy level to Product.
9. In the Attributes pane, select all the attributes by clicking the first one, and then holding Shift and
clicking the last one. In the Properties pane, set the AttributeHierarchyVisible property to False.
This hides the individual attributes, making the Categorized Products hierarchy the only way to
browse the Product dimension.
10. On the Dimension menu, click Process. If you are prompted to build and deploy the project first,
click Yes. If you are prompted to enter a password, type Pa55w.rd and click OK.
11. In the Process Dimension-Product dialog box, click Run and wait for the dimension to be
processed. Click Close, and click Close again to close the Process Dimension-Product dialog box.

12. On the Browser tab, ensure that the Categorized Products hierarchy is selected and expand the All
level to display the categories.

13. Expand the categories and subcategories to view individual products.

14. On the File menu, click Save All.

 MCT USE ONLY. STUDENT USE PROHIBITED
L3-3

 Task 2: Create a Non-Natural Hierarchy

1. In Solution Explorer, double-click the Customer.dim dimension to open it.

2. On the Dimension Structure tab, in the Attributes pane, drag Gender into an empty area of the
Hierarchies pane. This creates a new hierarchy named Hierarchy.

3. Right-click Hierarchy and click Rename, then rename the hierarchy to Gender-Marital Status.
4. In the Attributes pane, drag Marital Status to the <new level> area beneath Gender in the Gender-
Marital Status hierarchy.

Note: if a warning is displayed, saying that attribute relationships do not exist and that performance
may be decreased, ignore it. You will see how to use attribute relationships to optimize a hierarchy
later in this lab.

5. On the Dimension menu, click Process. If you are prompted to build and deploy the project first,
click Yes, and if you are prompted to confirm the database will be overwritten, click Yes.

6. In the Process Dimension- Customer dialog box, click Run and wait for the dimension to be
processed. Click Close, and click Close again to close the Process Dimension-Customer dialog box.
7. On the Browser tab, ensure that the Gender-Marital Status hierarchy is selected and expand the All
level to display the gender level. Note that the gender can be F, M, or Unknown.

8. Expand F and M, and note that the marital status level value can be M or S.

9. On the File menu, click Save All.

 Task 3: Create a Hierarchy with Attribute Relationships

Configure Attribute Column Bindings

1. In Solution Explorer, double-click the Date.dim dimension to open it in the Dimension Designer.
2. In the Properties pane, note that the Type property for this dimension has been set to Time. This
occurred when you added dimension intelligence in the previous exercise.

3. In the Attributes pane, right-click Calendar Semester, and click Properties.

4. In the Properties pane, scroll down to the Source section, click the KeyColumns field, and then click
the ellipses (…) button.

5. In the Key Columns dialog box, in the Available Columns table, click CalendarYear, and then click
the right arrow (>) icon.

6. Click the up arrow icon to move CalendarYear above CalendarSemester, and then click OK.

7. In the Properties pane, click the NameColumn field, and then click the ellipses (…) button.
8. In the Name Column dialog box, in the Source column field, click CalendarSemester, and then click
OK.

9. In the Properties pane, click the ValueColumn field, and then click the ellipses (…) button.

10. In the Value Column dialog box, in the Source column field, click CalendarSemester, and then click
OK.

11. In the Attributes pane, click Calendar Quarter.

12. In the Properties pane, scroll down to the Source section, click the KeyColumns field, and then click
the ellipses (…) button.

13. In the Key Columns dialog box, in the Available Columns table, click CalendarYear, and then click
the right arrow (>) icon.
 MCT USE ONLY. STUDENT USE PROHIBITED
L3-4 Developing SQL Data Models

14. Click the up arrow icon to move CalendarYear above CalendarQuarter, and then click OK.

15. In the Properties pane, click the NameColumn field, and then click the ellipses (…) button.

16. In the Name Column dialog box, in the Source column field, click CalendarQuarter, and then click
OK.

17. In the Properties pane, click the ValueColumn field, and then click the ellipses (…) button.

18. In the Value Column dialog box, in the Source column field, click CalendarQuarter, and then click
OK.

19. In the Attributes pane, click Month, and in the Properties pane, scroll down to the Source section,
click the KeyColumns field, and then click the ellipses (…) button.

20. In the Key Columns dialog box, in the Key Columns table, click EnglishMonthName, and then click
the left arrow (<) icon.
21. In the Available Columns table, click CalendarYear, and then click the right arrow (>) icon.

22. In the Available Columns table, click MonthNumberOfYear, then click the right arrow (>) icon.
Ensure that CalendarYear appears above MonthNumberOfYear, and then click OK.

23. In the Properties pane, click the NameColumn field, and then click the ellipses (…) button.

24. In the Name Column dialog box, in the Source column field, click EnglishMonthName, and then
click OK.
25. In the Properties pane, click the ValueColumn field, and then click the ellipses (…) button.

26. In the Value Column dialog box, in the Source column field, click EnglishMonthName, and then
click OK.
27. On the File menu, click Save All.

Create Attribute Relationships

1. In the Date.dim Dimension Designer, click the Attributes Relationships tab.

2. In the diagram pane, right-click an empty space, and then click New Attribute Relationship.

3. In the Create Attribute Relationship dialog box, in the Source Attribute section, in the Name field,
click Full Date Alternate Key. Then, in the Related Attribute section, in the Name field, click
Month.

4. In the Relationship type field click Rigid (will not change over time), because a particular date will
always be in the same month, and then click OK.

5. Repeat the previous three steps to create the following relationships:

o Month > Calendar Quarter (Rigid)

o Calendar Quarter > Calendar Semester (Rigid)

o Calendar Semester > Calendar Year (Rigid)

6. On the File menu, click Save All.

Create a Hierarchy

1. In the Date.dim Dimension Designer, on the Dimension Structure tab, in the Attributes pane, drag
Calendar Year into the Hierarchies pane. A new hierarchy named Hierarchy is created.

2. Right-click Hierarchy and click Rename then rename the hierarchy to Calendar Date.
 MCT USE ONLY. STUDENT USE PROHIBITED
L3-5

3. In the Attributes pane, drag the following attributes one-by-one to the <new level> area in the
Calendar Date hierarchy:
o Calendar Semester

o Calendar Quarter

o Month

o Full Date Alternate Key

4. In the Calendar Date hierarchy, right-click Full Date Alternate Key and click Rename. Then rename
the hierarchy level to Day.
5. In the Attributes pane, select Calendar Year, press the Ctrl key, and click Calendar Semester,
Calendar Quarter, Month, and Full Date Alternate Key to select all of these attributes.
6. In the Properties pane, change the AttributeHierarchyVisible property to False. This defines these
attributes as member properties rather than hierarchies in their own right, so that users can only
browse them through the Calendar Date hierarchy.

7. On the Dimension menu, click Process. If you are prompted to build and deploy the project first,
click Yes, and if you are prompted to overwrite the database click Yes.

8. In the Process Dimension-Date dialog box, click Run and wait for the dimension to be processed.
Click Close, and click Close again to close the Process Dimension-Date dialog box.

Results: After this exercise, you should have created a Categorized Products hierarchy and a Gender-
Marital Status hierarchy.

Exercise 3: Sorting and Grouping Dimension Attributes

 Task 1: Sort Attribute Members
1. In the Browser tab, in the Hierarchy drop-down, ensure Calendar Date is selected.
2. Expand All, expand 2005, and then expand 1. Note that months are displayed in alphabetical order
instead of chronological order.
3. On the Dimension Structure tab, in the Attributes pane, right-click the Month attribute and click
Properties.

4. In the Properties pane, in the OrderBy box, select Key.

5. On the Dimension menu, click Process. If you are prompted to build and deploy the project first,
click Yes. If you are prompted to overwrite the database, click Yes. In the Process Dimension-Date
dialog, click Run.

6. When processing is complete, click Close and then, in the Process Dimension-Date dialog, click
Close.

7. In the Browser tab, in the Hierarchy drop-down, ensure Calendar Date is selected.

8. In the Dimension menu, click Refresh. Note that the months are now correctly sorted in
chronological order.
 MCT USE ONLY. STUDENT USE PROHIBITED
L3-6 Developing SQL Data Models

 Task 2: Group Attribute Members

1. In Solution Explorer, double-click the Customer.dim dimension to open it in the Dimension Designer.

2. On the Dimension Structure tab, in the Data Source View pane, right-click the Customer table, and
then click Explore Data. Notice the range of values for the YearlyIncome column, and then close the
Explore Customer Table window.

3. On the Browser tab, change the Hierarchy field to Yearly Income.

4. Expand All, and notice that the data is unstructured.

5. On the Dimension Structure tab, in the Attributes pane, click Yearly Income.
6. In the Properties pane, in the DiscretizationMethod box, click Automatic, in the
DiscretizationBucketCount box type 5 and, in the OrderBy list, select Key.

7. On the Dimension menu, click Process. If you are prompted to build and deploy the project first,
click Yes, and if you are prompted to overwrite the database, click Yes.
8. In the Process Dimension-Customer dialog, click Run and then, when processing is complete, click
Close. Click Close in the Process Dimension-Customer dialog.
9. On the Browser tab, click Reconnect and verify that the yearly income is grouped into five ranges,
with a sixth member for unknown values.

Results: You will have a multidimensional project with enhanced dimensions.

 MCT USE ONLY. STUDENT USE PROHIBITED
L4-1

Module 4: Working with Measures and Measure Groups

Lab: Configuring Measures and Measure
Groups
Exercise 1: Configuring Measures
 Task 1: Prepare the Lab Environment
1. Ensure that the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and then
log on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd.

2. In the D:\Labfiles\Lab04\Starter folder, right-click Setup.cmd, and then click Run as

administrator.

3. Click Yes when prompted to confirm that you want to run the command file, and then wait for the
script to finish.

 Task 2: Create a Measure Group

1. Start SQL Server Data Tools and open the Adventure Works OLAP.sln solution in the
D:\Labfiles\Lab04\Starter folder.

2. In Solution Explorer, right-click Adventure Works DSV.dsv, and then click View Designer.

3. On the Data Source View menu, click Add/Remove Tables.

4. In the Add/Remove Tables dialog box, in the Available objects list, select FactResellerSales (dbo),
and then click the > button to add the selected table to the Included objects list.

5. Click Add Related Tables to add the tables that are related to FactResellerSales, and then click OK.
Note that the tables have been added to the data source view, and then, on the File menu, click Save
All.

6. Close the Adventure Works DSV.dsv designer.

7. In Solution Explorer, right-click Sales.cube, and then click View Designer.

8. On the Cube menu, click New Measure Group, and then in the New Measure Group dialog box,
select FactResellerSales, and then click OK.

9. In the Measures pane, right-click the Fact Reseller Sales measure group, click Rename, and then
change the name of the measure group to Reseller Sales.

10. Expand the Reseller Sales measure group, and then review the names of the measures that it
contains. Note that when the Reseller Sales measure group was set up, measures were created for all
of the numerical fields in the FactResellerSales table.

 Task 3: Modify Measure Groups

1. On the Cube menu, point to Show Measures In, and then click Grid to view all of the measures in
the cube as a grid. Note the aggregations that are used to summarize the measures when they are
analyzed across dimensions.

2. Click the row for the Revision Number measure, and then hold down the Ctrl key and click the rows
for the following measures to select them all:

o Unit Price

o Extended Amount
 MCT USE ONLY. STUDENT USE PROHIBITED
L4-2 Developing SQL Data Models

o Unit Price Discount Pct

o Discount Amount

o Product Standard Cost

o Tax Amt

o Freight

3. Click the Delete button, and then in the Delete Objects dialog box, click OK to remove these
measures.

4. Right-click the Order Quantity measure, click Rename, and then change the name of the measure to
Reseller Order Quantity.

5. Repeat the previous step to rename the following measures:

o Total Product Cost (rename to Reseller Cost).

o Sales Amount (rename to Reseller Revenue).

o Fact Reseller Sales Count (rename to Reseller Sales Count).

6. On the Cube menu, point to Show Measures In, click Tree to view measures in the cube as a tree,
and then, on the File menu, click Save All.

Results: After this exercise, you should have created a new measure group for the FactResellerSales
table, removed unrequired measures, and renamed measures.

Exercise 2: Defining a Regular Relationship

 Task 1: View Existing Dimensions for Measure Groups
1. On the Build menu, click Deploy Solution. If you are prompted for an account password, type
Pa55w.rd, and then click OK.

2. Wait for the Deployment Completed Successfully message in the Deployment Progress –
Adventure Works OLAP window, and then close the Deployment Progress – Adventure Works OLAP
window.

3. In Solution Explorer, right-click Sales.cube, and then click Browse.

4. In the Metadata pane, in the Measure Group drop-down list, select Internet Sales. Notice that
there is a Customer dimension.

5. In the Measure Group drop-down list, select Reseller Sales and notice that there is no Customer
dimension, because reseller sales are sold to resellers, which are defined in a different dimension
table.

 Task 2: Create a Dimension

1. In Solution Explorer, right-click Dimensions, and then click New Dimension.
2. In the Dimension Wizard, on the welcome page, click Next.

3. On the Select Creation Method page, ensure that Use an existing table is selected, and then click
Next.
 MCT USE ONLY. STUDENT USE PROHIBITED
L4-3

4. On the Specify Source Information page, in the Main table list, select DimReseller, and then click
Next.
5. On the Select Related Tables page, clear Geography and Sales Territory, and then click Next.

6. On the Select Dimension Attributes page, select only the following attributes, and then click Next:

o Reseller Key

o Business Type

o Reseller Name

7. On the Completing the Wizard page, in the Name box, type Reseller, and then click Finish.
8. In the Sales.cube designer, click the Dimension Usage tab.

9. On the Cube menu, click Add Cube Dimension.

10. In the Add Cube Dimension dialog box, select Reseller, and then click OK.

11. On the Dimension Usage tab, click the Reseller Key cell at the intersection of the Reseller Sales
measure group and the Reseller dimension, and then click the ellipsis (…) button.
12. In the Define Relationship dialog box, note that a Regular relationship type has been detected, and
then click Cancel.

13. On the File menu, click Save All.

14. On the Build menu, click Deploy Solution. If you are prompted for an account password, type
Pa55w.rd, and then click OK.

15. Wait for the Deployment Completed Successfully message in the Deployment Progress –
Adventure Works OLAP window, and then close the Deployment Progress – Adventure Works OLAP
window.

16. In Solution Explorer, right-click Sales.cube, and then click Browse.

17. On the Cube menu, click Reconnect.

18. In the Metadata pane, in the Measure Group drop-down list, select Reseller Sales.

19. Click the Click to execute the query link, notice that there is now a Reseller dimension.

20. Expand Measures, expand Reseller Sales, and then drag Reseller Revenue to the query results area.
21. Expand Reseller, and then drag Business Type to the left of the Reseller Revenue column.

22. Click the Click to execute the query link to see reseller sales revenue broken down by business type.

Results: After this exercise, you should have added a Reseller dimension that uses a regular relationship
with the Reseller Sales measure group to enable you to analyze reseller sales data.

Exercise 3: Configuring Measure Group Storage

 Task 1: Configure Proactive Caching
1. In the Cube Designer for the Sales cube, on the Cube Structure tab, in the Measures pane, right-
click Internet Sales, and then click Properties.
2. In the Properties window, click ProactiveCaching, and then click the ellipsis (…) button.
 MCT USE ONLY. STUDENT USE PROHIBITED
L4-4 Developing SQL Data Models

3. Under Standard setting, review each of the storage setting options. When you have finished, drag
the slider to Automatic MOLAP, and then click OK.
4. Repeat the previous steps to configure the Reseller Sales measure group for Automatic MOLAP
proactive caching.

 Task 2: Design Aggregations

1. In the Cube Designer for the Sales cube, click the Aggregations tab.

2. Right-click Internet Sales (0 Aggregation Designs), and then click Design Aggregations.

3. On the Welcome to the Aggregation Design Wizard page, click Next.

4. On the Review Aggregation Usage page, click Set All to Default, and then click Next.
5. On the Specify Object Counts page, click Count. When the count process has completed, click Next.

6. On the Set Aggregations Options page, select Performance gain reaches, ensure that the value is
set to 30%, and then click Start.

7. When the wizard has designed aggregations for a performance gain of at least 30 percent, click Next.

8. On the Completing the Wizard page, change the name of the aggregation to InternetSalesAgg,
select Save the aggregations but do not process them, and then click Finish.

9. Repeat the previous steps to design Reseller Sales aggregations for a performance gain of 30
percent. Name the aggregation design ResellerSalesAgg.

10. On the File menu, click Save All.

11. On the Build menu, click Deploy Solution. If you are prompted for an account password, type
Pa55w.rd, and then click OK.
12. Wait for the Deployment Completed Successfully message in the Deployment Progress –
Adventure Works OLAP window, and then close the Deployment Progress – Adventure Works OLAP
window.

13. Close SQL Server Data Tools.

14. Start SQL Server Management Studio. When you are prompted, in the Connect to Server dialog
box, specify the following settings, and then click Connect:

o Server type: Analysis Services

o Server name: localhost

15. In Object Explorer, expand Databases, expand Adventure Works OLAP, expand Cubes, expand
Sales, expand Measure Groups, expand Internet Sales, and then expand Aggregation Designs.

16. Verify that the InternetSalesAgg aggregation design has been deployed with the cube.

17. Expand Reseller Sales and its Aggregation Designs folder to verify that the ResellerSalesAgg
aggregation design has been deployed with the cube.
18. Close SQL Server Management Studio.

Results: After this exercise, you should have configured proactive caching and defined the storage mode
aggregations for the Internet Sales and Reseller Sales measure groups.
 MCT USE ONLY. STUDENT USE PROHIBITED
L5-1

Module 5: Introduction to MDX

Lab: Using MDX
Exercise 1: Adding Calculated Members
 Task 1: Prepare the Lab Environment
1. Ensure that the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and then
log on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd.

2. In the D:\Labfiles\Lab05\Starter folder, right-click Setup.cmd, and then click Run as

administrator.

3. Click Yes when prompted to confirm that you want to run the command file, and then wait for the
script to finish.

 Task 2: Add a Calculated Member by Using Form View

1. On the Start screen, type SQL Server Data Tools 2015, and then press Enter.
2. On the File menu, point to Open, click Project/solution, browse to the D:\Labfiles\Lab05\Starter
folder, and then double-click Adventure Works OLAP.sln.

3. In Solution Explorer, right-click Sales.cube, and then click View Designer.

4. On the Cube menu, point to View, and then click Calculations.

5. On the Cube menu, click New Calculated Member.

6. In the Name text box, type [Total Revenue].

7. In the Parent hierarchy box, ensure that Measures is selected.

8. In the Expression text box, type the following MDX expression:

[Measures].[Internet Revenue] + [Measures].[Reseller Revenue]

9. In the Format string drop-down list, select "Currency".

10. In the Non-empty behavior list, select Internet Revenue and Reseller Revenue, and then click OK.

11. On the File menu, click Save Selected Items.

12. Above the form, click the Script View button, and then check the MDX text that has been added to
the cube.

 Task 3: Add Calculated Members by Using Script View

1. In the Script Editor pane, place the cursor after the last semicolon, and then press Enter.

2. Type the following code:

CREATE MEMBER CURRENTCUBE.[Measures].[Internet Profit]

AS ([Measures].[Internet Revenue] - [Measures].[Internet Cost]) /
[Measures].[Internet Cost],
FORMAT_STRING = "Percent",
NON_EMPTY_BEHAVIOR = { [Internet Revenue], [Internet Cost] },
VISIBLE = 1,
ASSOCIATED_MEASURE_GROUP = 'Internet Sales' ;
 MCT USE ONLY. STUDENT USE PROHIBITED
L5-2 Developing SQL Data Models

3. Press Enter, and then type the following code:

CREATE MEMBER CURRENTCUBE.[Measures].[Reseller Profit]

AS ([Measures].[Reseller Revenue] - [Measures].[Reseller Cost]) /
[Measures].[Reseller Cost],
FORMAT_STRING = "Percent",
NON_EMPTY_BEHAVIOR = { [Reseller Revenue], [Reseller Cost] },
VISIBLE = 1,
ASSOCIATED_MEASURE_GROUP = 'Reseller Sales' ;

4. On the File menu, click Save Selected Items.

5. On the Build menu, click Deploy Solution. If you are prompted to overwrite the database, click Yes.
If you are prompted for impersonation credentials, enter the password Pa55w.rd for
ADVENTUREWORKS\ServiceAcct, and then click OK.

6. When the deployment has completed successfully, close Visual Studio.

Results: After this exercise, you should have added three calculated members to the Sales cube and
deployed the cube to the MIA-SQL instance of SQL Server Analysis Services.

Exercise 2: Querying a Cube by Using MDX

 Task 1: Test Total Revenue in SQL Server Management Studio
1. Start SQL Server Management Studio.

2. In the Connect to Server dialog box, in the Server type drop-down list, select Analysis Services.

3. In the Server name drop-down list, select MIA-SQL, and then click Connect.

4. In Object Explorer, expand Databases, right-click Adventure Works OLAP, point to New Query,
and then click MDX.

5. In the query pane, type the following code:

SELECT
{
[Measures].[Internet Revenue],
[Measures].[Reseller Revenue],
[Measures].[Total Revenue]
} ON COLUMNS,
[Product].[Categorized Products].Members ON ROWS
FROM [Sales]

6. Click Execute, and then examine the results.

 MCT USE ONLY. STUDENT USE PROHIBITED
L5-3

 Task 2: Test Profit Calculations in SQL Server Management Studio

1. In SQL Server Management Studio, in Object Explorer, right-click Adventure Works OLAP, point to
New Query, and then click MDX.

2. In the query pane, type the following code:

SELECT
{
[Measures].[Internet Cost],
[Measures].[Internet Revenue],
[Measures].[Internet Profit]
} ON COLUMNS,
[Customer].[City].Members ON ROWS
FROM [Sales]

3. Click Execute, and then examine the results.

4. In the query pane, after the existing code, add the following line of code:

WHERE ([Customer].[State Province Name].&[New York])

5. Click Execute, and then examine the results.

6. In SQL Server Management Studio, in Object Explorer, right-click Adventure Works OLAP, point to
New Query, and then click MDX.

7. In the query pane, type the following code:

SELECT
{
[Measures].[Reseller Cost],
[Measures].[Reseller Revenue],
[Measures].[Reseller Profit]
} ON COLUMNS,
[Sales Territory].[Sales Territory].Members ON ROWS
FROM [Sales]

8. Click Execute, and then examine the results.

9. Close SQL Server Management Studio, without saving any changes.

 Task 3: Use Excel to Execute MDX Queries

1. On the Start menu, type Excel 2016, and then press Enter.
2. On the home page, click Blank workbook.

3. On the ribbon, click the Data tab.

4. In the Get & Transform group, click New Query, point to From Database, and then click From SQL
Server Analysis Services Database.

5. In the Microsoft SQL Server Analysis Services Database dialog box, in the Server box, type MIA-
SQL, and click OK.

6. Then click Connect.

7. In the Navigator pane, expand Adventure Works OLAP, expand the Sales folder, expand the Sales
cube, and then expand the Internet Sales measure group.

8. Select Internet Cost, Internet Revenue, and Internet Profit.

9. Expand Product, and then expand Categorized Products.

 MCT USE ONLY. STUDENT USE PROHIBITED
L5-4 Developing SQL Data Models

10. Select Category and Subcategory, and then click Load.

 Task 4: Use an MDX Query in a PivotTable

1. In Excel, click Sheet 1.

2. On the ribbon, click the Data tab.

3. Click Get External Data, click From Other Sources, and then click From Analysis Services.
4. In the Data Connection Wizard dialog box, on the Connect to Database Server page, in the Server
name text box, type MIA-SQL, and then click Next.

5. On the Select Database and Table page, ensure that the Adventure Works OLAP database and the
Sales cube are selected, and then click Next.
6. On the Save Data Connection File and Finish page, click Finish.

7. In the Import Data dialog box, click New worksheet, and then click OK.

8. In the PivotTable Fields pane, select Internet Revenue, Reseller Revenue, and Total Revenue.

9. Under Sales Territory, select Sales Territory.

10. In the worksheet, expand Europe, and note the Total Revenue value for the United Kingdom.

11. Close Excel, without saving any changes.

Results: After this exercise, you should have created several test MDX queries in SQL Server Management
Studio and an Excel spreadsheet that connects to the Sales OLAP cube.
 MCT USE ONLY. STUDENT USE PROHIBITED
L6-1

Module 6: Customizing Cube Functionality

Lab: Customizing a Cube
Exercise 1: Implementing an Action
 Task 1: Prepare the Lab Environment
1. Ensure that the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and then
log on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd.

2. In the D:\Labfiles\Lab06\Starter folder, right-click Setup.cmd, and then click Run as

administrator.

3. Click Yes when prompted to confirm that you want to run the command file, and then wait for the
script to finish.

 Task 2: Create a Drill-Through Action

1. Start Visual Studio, on the File menu, point to Open, and then click Project/Solution. Browse to
D:\Labfiles\Lab06\Starter folder and open the Adventure Works OLAP.sln solution.

2. On the Build menu, click Deploy Solution.

3. If the Microsoft Visual Studio dialog box appears, click Yes.

4. If you are prompted for impersonation credentials, enter the password Pa55w.rd for
ADVENTUREWORKS\ServiceAcct, and then click OK.

5. When deployment is complete, close the Deployment Progress window.

6. In Solution Explorer, double-click Sales.cube to open it in the Cube Designer.

7. In the Cube Designer, click the Actions tab.

8. On the Cube menu, click New Drillthrough Action.

9. In the Name box, change the name of this action to Internet Sales Details.
10. In the Measure group members list, select Internet Sales.

11. In the Drillthrough Columns box, in the Select dimensions list, select Customer, in the Return
Columns list, select City and Full Name, and then click OK.

12. In the Drillthrough Columns box, in the Select dimensions list, click Order Date, in the Return
Columns list, select Full Date Alternate Key, and then click OK.

13. In the Drillthrough Columns box, in the Select dimensions list, click Product, in the Return
Columns list, select English Product Name, and then click OK.

14. Expand Additional Properties, and then in the Caption box, type Drillthrough to Order Details.

15. On the File menu, click Save All.

 Task 3: Browse a Drill-Through Action

1. In Visual Studio, on the Build menu, click Deploy Solution. If you are prompted for impersonation
credentials, enter the password Pa55w.rd for ADVENTUREWORKS\ServiceAcct, and then click OK.

2. Close the Deployment Progress window.

3. In the Cube Designer, click the Browser tab.

4. On the Cube menu, click Analyze in Excel. If a security notice is displayed, click Enable.
 MCT USE ONLY. STUDENT USE PROHIBITED
L6-2 Developing SQL Data Models

5. In Excel, in the PivotTable Fields pane, under Internet Sales, select Internet Revenue.

6. In the PivotTable Fields pane, under Order Date, select Order Date.Calendar Date.

7. In the PivotTable Fields pane, under Product, select Categorized Products.

8. In the PivotTable, right-click the sales amount for Bikes in 2007, point to Additional Actions, and
then click Drillthrough to Order Details.

9. View the new worksheet that is generated by the drill-through action, noting that it shows the city,
customer name, and product name for each sale of a bike in 2007.

10. Close Excel without saving the workbook.

11. Keep Visual Studio open for the next exercise.

Results: After this exercise, you should have a new drill-through action in the solution.

Exercise 2: Implementing Perspectives

 Task 1: Create a Perspective
1. In Visual Studio, in the Sales.cube Cube Designer, click the Perspectives tab.
2. On the Cube menu, click New Perspective.

3. Clear the first row of the first Perspective Name column, and then type Internet Sales.

4. In the same column, clear the check box for the following objects:
1. The Reseller Sales measure group (this clears all of the measures in this measure group).

2. The Reseller dimension.

3. The Reseller Profit calculated member.

4. On the Cube menu, click New Perspective.

5. Clear the first row of the second Perspective Name column, and then type Reseller Sales.

6. In the same column, clear the check box for the following objects:

a. The Internet Sales measure group (this clears all of the measures in this measure group).

b. The Customer dimension.

c. The Internet Sales Details action.

d. The Internet Profit calculated member.

7. On the File menu, click Save All.

 Task 2: Browse a Perspective

1. In Visual Studio, on the Build menu, click Deploy Solution. If you are prompted for impersonation
credentials, enter the password Pa55w.rd for ADVENTUREWORKS\ServiceAcct, and then click OK.

2. When deployment has successfully completed, close the Deployment Progress window.

3. In the Cube Designer, click the Browser tab, and then click the Reconnect icon.

4. In the cube selection area above the Metadata pane, click the ellipsis (…), select Internet Sales, and
then click OK.
 MCT USE ONLY. STUDENT USE PROHIBITED
L6-3

5. In the Metadata pane, expand Measures, and then note that only the Internet Sales measure group
is shown in this perspective.

6. Expand Internet Sales, and then drag the Internet Revenue measure to the query results area.

7. In the Metadata pane, expand the Customer dimension, and then drag the Yearly Income hierarchy
to the left of the Internet Revenue figure.

8. Click the Click to execute the query link.

9. This displays Internet revenue for customers, broken down by income band.

10. On the Cube menu, click Analyze in Excel.

11. In the Analyze in Excel dialog box, select the Reseller Sales perspective, and then click OK. If a
security notice is displayed, click Enable.

12. In Excel, in the PivotTable Fields pane, note that only the Reseller Sales measures are shown.

13. In the PivotTable Fields pane, under Reseller Sales, select Reseller Revenue.

14. In the PivotTable Fields pane, under Reseller, select Business Type. The PivotTable shows revenue
for reseller sales, broken down by business type.

15. Close Excel without saving the workbook.

Results: After this exercise, you should have defined a new perspective in the solution.

Exercise 3: Implementing a Translation

 Task 1: Create Dimension Translations
1. In Visual Studio, in Solution Explorer, in the Dimensions folder, double-click Date.dim to open it in
the dimension designer, and then click the Translations tab.

2. On the Dimension menu, click New Translation.

3. In the Select Language dialog box, select French (France), and then click OK.
4. In the row for the Calendar Date hierarchy, type Date du Calendrier in the French (France)
translation column.

5. In the row for the Calendar Year level, type Année in the French (France) translation column.

6. Note: To type é, press Alt+130 by using the number pad. Ensure that Num Lock is on. If this
does not work, type the captions without accents.

7. In the row for the Calendar Semester level, type Semestre in the French (France) translation column.

8. In the row for the Calendar Quarter level, type Trimestre in the French (France) translation column.

9. In the row for the Month level, type Mois in the French (France) translation column.

10. In the row for the Day level, type Journée in the French (France) translation column.

11. On the Translation tab toolbar, click Show All Attributes. This reveals the attributes that are hidden
in the dimension.

12. Click the French (France) translation row for the Month attribute, and then click the [.] button.

13. In the Attribute Data Translation dialog box, in the Translation columns list, click
FrenchMonthName, and then click OK.
 MCT USE ONLY. STUDENT USE PROHIBITED
L6-4 Developing SQL Data Models

14. On the File menu, click Save All.

15. Close the Date.dim dimension designer.

 Task 2: Create Cube Translations

1. In Solution Explorer, double-click Sales.cube to view it in the Cube Designer.

2. In the Cube Designer, click the Translations tab.

3. On the Cube menu, click New Translation, and then, in the Select Language dialog box, click
French (France), and then click OK.

4. In the row for the Internet Sales measure group, in the French (France) translation column, type
Ventes d'Internet.
5. In the row for the Internet Revenue measure, in the French (France) translation column, type
Revenu d’Internet.
6. In the row for the Order Date dimension, in the French (France) translation column, type Date de
Vente.

7. On the File menu, click Save All.

 Task 3: Browse Translations

1. In Visual Studio, on the Build menu, click Deploy Solution. If you are prompted for impersonation
credentials, enter the password Pa55w.rd for ADVENTUREWORKS\ServiceAcct, and then click OK.

2. When deployment is complete, close the Deployment Progress window.

3. In the Cube Designer, click the Browser tab, and then click the Reconnect icon.
4. On the toolbar, in the Language list, select French (France).

5. In the Metadata pane, expand Measures, expand Ventes d’Internet, and then drag Revenu
d’Internet to the query results area.
6. In the Metadata pane, expand the Date de Vente dimension, and then drag the Date de
Vente.Date du Calendrier hierarchy to the left of Revenu d’Internet.

7. Click the Click to execute the query link. Note that the captions are displayed in French, and that
the month names in the Mois column are also in French.

8. Close Visual Studio.

Results: After this exercise, you should have French data and metadata in the cube.
 MCT USE ONLY. STUDENT USE PROHIBITED
L7-1

Module 7: Implementing a Tabular Data Model by Using SQL

Server Analysis Services
Lab: Implementing a Tabular Data Model in
SQL Server Analysis Services
Exercise 1: Creating a Tabular Data Model Project in SQL Server Analysis
Services
 Task 1: Prepare the Lab Environment
1. Ensure that the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and then
log on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd.

2. In the D:\Labfiles\Lab07\Starter folder, right-click Setup.cmd, and then click Run as

administrator.

3. Click Yes when prompted to confirm that you want to run the command file, and then wait for the
script to finish.

 Task 2: Create a Tabular Project in SQL Server Analysis Services

1. Start SQL Server Data Tools 2015.

2. On the File menu, point to New, and then click Project.

3. In the New Project dialog box, type the following values, and then click OK:

o Project Template: Analysis Services Tabular Project

o Name: AWSalesTab
o Location: D:\Labfiles\Lab07\Starter

4. If the Tabular model designer dialog box is displayed, in the Workspace server list, select
localhost\SQL2, in the Compatibility level box, select SQL Server 2017 / Azure Analysis Services
(1400), and then click OK.

 Task 3: Import Tables into the Data Model

1. On the Model menu, click Import from Data Source.

2. In the Get Data dialog box, select SQL Server database, and then click Connect.

3. On the SQL Server database page, type the following values, and then click OK:

o Server name: MIA-SQL

o Database name: AdventureWorksDW

4. On the Impersonation Information page, in the Impersonate Mode box, select Impersonate
Service Account, and then click Connect.

5. In the Encryption Support dialog box, click OK.

6. On the Navigator page, select the following source tables, then click OK:

o DimDate

o DimEmployee

o DimGeography
 MCT USE ONLY. STUDENT USE PROHIBITED
L7-2 Developing SQL Data Models

o DimProduct

o DimProductCategory

o DimProductSubcategory

o DimReseller

o FactResellerSales

7. In the Query Editor, in the Queries pane, select DimDate.

8. In the Query Settings pane, in the Name box, change DimDate to Date, then press enter.

9. On the Columns menu, click Choose Columns.

10. In the Choose Columns dialog box, clear the following columns, and then click OK:

o SpanishDayNameOfWeek

o FrenchDayNameOfWeek

o DayNumberOfYear

o WeekNumberOfYear

o SpanishMonthName
o FrenchMonthName

o CalendarSemester

o FiscalSemester
11. In the Queries pane, select DimEmployee.

12. In the Query Settings pane, in the Name box, change DimEmployee to Employee, then press Enter.

13. On the Columns menu, click Choose Columns.

14. In the Choose Columns dialog box, clear the following columns, and then click OK:

o SalesTerritoryKey

o NameStyle

o Title

o HireDate

o BirthDate

o LoginID

o EmailAddress

o Phone

o MaritalStatus

o EmergencyContactName

o EmergencyContactPhone
o SalariedFlag

o Gender

o PayFrequency
o Baserate
 MCT USE ONLY. STUDENT USE PROHIBITED
L7-3

o VacationHours

o SickLeaveHours

o CurrentFlag

o SalesPersonFlag

o StartDate

o EndDate

o Status

15. In the Queries pane, select DimGeography.

16. In the Query Settings pane, in the Name box, change DimGeography to Geography, then press
Enter.

17. On the Columns menu, click Choose Columns.

18. In the Choose Columns dialog box, clear the following columns, and then click OK:

o SpanishCountryRegionName

o FrenchCountryRegionName

o IpAddressLocator

19. In the Queries pane, select DimProduct.

20. In the Query Settings pane, in the Name box, change DimProduct to Product, then press Enter.

21. On the Columns menu, click Choose Columns.

22. In the Choose Columns dialog box, clear the following columns, and then click OK:

o WeightUnitMeasureCode
o SizeUnitMeasureCode

o SpanishProductName

o FrenchProductName
o FinishedGoodsFlag

o SafetyStockLevel

o ReorderPoint

o DaysToManufacture

o ProductLine

o DealerPrice

o Class

o Style

o ModelName
o FrenchDescription

o ChineseDescription

o ArabicDescription

o HebrewDescription
 MCT USE ONLY. STUDENT USE PROHIBITED
L7-4 Developing SQL Data Models

o ThaiDescription

o GermanDescription

o JapaneseDescription

o TurkishDescription

o StartDate

o EndDate

o Status

23. In the Queries pane, select DimProductCategory.

24. In the Query Settings pane, in the Name box, change DimProductCategory to Product Category,
then press Enter.

25. On the Columns menu, click Choose Columns.

26. In the Choose Columns dialog box, clear the following columns, and then click OK:

o SpanishProductCategoryName

o FrenchProductCategoryName

27. In the Queries pane, select DimProductSubcategory.

28. In the Query Settings pane, in the Name box, change DimProductSubcategory to Product
Subcategory, then press Enter.

29. On the Columns menu, click Choose Columns.

30. In the Choose Columns dialog box, clear the following columns, and then click OK:

o SpanishProductSubcategoryName

o FrenchProductSubcategoryName

31. In the Queries pane, select DimReseller.

32. In the Query Settings pane, in the Name box, change DimReseller to Reseller, then press enter.
33. On the Columns menu, click Choose Columns.

34. In the Choose Columns dialog box, clear the following columns, and then click OK:

o OrderFrequency

o OrderMonth

o FirstOrderYear

o LastOrderYear

o ProductLine

o AddressLine1

o AddressLine2
o AnnualSales

o BankName

o MinPaymentType

o MinPaymentAmount
 MCT USE ONLY. STUDENT USE PROHIBITED
L7-5

o AnnualRevenue

o YearOpened

35. In the Queries pane, select FactResellerSales.

36. In the Query Settings pane, in the Name box, change FactResellerSales to Reseller Sales, then
press enter.

37. On the Columns menu, click Choose Columns.

38. In the Choose Columns dialog box, clear the following columns, and then click OK:

o DueDateKey

o PromotionKey

o CurrencyKey

o SalesTerritoryKey
o RevisionNumber

o CarrierTrackingNumber

o CustomerPONumber

o DueDate

39. On the toolbar, click Import, and then wait for the data to be imported. When the data has been
imported successfully, click Close.

 Task 4: Create Measures

1. In the Model.bim pane, on the Reseller Sales tab, click the first empty cell in the grid under the
OrderQuantity column.

2. On the Column menu, point to AutoSum, click Sum, and then, in the formula bar, modify the
expression that has been generated to name the measure Quantity, as shown in the following
example:

Quantity:=Sum([OrderQuantity])

3. Widen the OrderQuantity column to see the calculated Quantity measure, and then click the cell in
the measure grid that contains the calculated value.

4. Press F4 to view the Properties pane, and then set the Format property to Whole Number, and the
Show Thousand Separator property to True.

5. Click the first empty cell under the TotalProductCost column.

6. On the Column menu, point to AutoSum, click Sum, and then modify the expression that is
generated to name the measure Cost, as shown in the following example:

Cost:=Sum([TotalProductCost])

7. Widen the TotalProductCost column to see the calculated Cost measure.

8. Select the cell that contains the calculated Cost measure, and then, in the Properties pane, ensure
that the Format property is set to Currency.

9. Click the first empty cell under the SalesAmount column.

 MCT USE ONLY. STUDENT USE PROHIBITED
L7-6 Developing SQL Data Models

10. On the Column menu, point to AutoSum, click Sum, and then modify the expression that is
generated to name the measure Revenue, as shown in the following example:

Revenue:=Sum([SalesAmount])

11. Widen the SalesAmount column to see the calculated Revenue measure.

12. Select the cell that contains the calculated Revenue measure, and then, in the Properties pane,
ensure that the Format property is set to Currency.

 Task 5: Test the Model

1. On the Model menu, click Analyze in Excel.

2. In the Analyze in Excel dialog box, ensure that Current Windows User is selected and that the
(Default) perspective is specified, and then click OK.

3. When Excel opens, in the PivotTable Fields pane, note that the measures that you defined are
displayed under a Reseller Sales measure group. The Reseller Sales table is also shown in this pane,
and contains the columns in the table on which the measures are based, which some users might find
confusing. You will fix this problem in the next exercise.

4. Select the Revenue measure, and then note that it is displayed in the PivotTable.

5. In the Geography table, select EnglishCountryRegionName. The revenue is shown for each country
or region, but the column name that is used to display this is not user-friendly.

6. Clear the EnglishCountryRegionName field.

7. In the Date table, select CalendarYear and EnglishMonthName. The revenue is shown for each
year, and each month. However, the months are listed in alphabetical rather than chronological order.
In addition, the Reseller Sales table contains two date keys (OrderDateKey and ShipDateKey), but it
is unclear which of these dates is used to determine the year and month. You will fix these problems
in the next exercise.

8. Close Excel without saving the workbook.

Results: After this exercise, you should have created a tabular data model project.

Exercise 2: Configuring Columns and Relationships

 Task 1: Configure Relationships
1. On the Model menu, point to Model View, and then click Diagram View. The tables are shown as a
schema diagram, with lines between them to denote relationships. Note that there are two
relationships between the Reseller Sales table and the Date table.

2. Double-click the solid line between the Reseller Sales and Date tables, note the columns that are
used to define the relationship, and that this is an active relationship, and then click Cancel.
3. Double-click the dotted line between the Reseller Sales and Date tables, note the columns that are
used to define the relationship, and that this is an inactive relationship, and then click Cancel.

4. If the inactive relationship that you noted above is based on the OrderDateKey column, right-click
the dotted line, and then click Mark as Active so that the active relationship is based on the
OrderDateKey column.
 MCT USE ONLY. STUDENT USE PROHIBITED
L7-7

5. Right-click the dotted relationship line (which should now represent the relationship that is based on
the ShipDateKey column), and then click Delete.

6. In the Confirm dialog box, click Delete from Model.

7. Right-click the Date table title bar, click Rename, and then rename the table to Order Date.
8. In the Tabular Model Explorer, expand Data Sources, right-click SQL/MIA-
SQL;AdventureWorksDW, and then click Import New Tables.

9. On the Navigator page, select DimDate, and then press OK.

10. In the Query Editor dialog box, in the Query Settings pane, in the Name box, change DimDate to
Ship Date, and then press Enter.

11. On the Columns menu, click Choose Columns.

12. In the Choose Columns dialog box, clear the following columns, and then click OK:

o SpanishDayNameOfWeek

o FrenchDayNameOfWeek

o DayNumberOfYear
o WeekNumberOfYear

o SpanishMonthName

o FrenchMonthName

o CalendarSemester

o FiscalSemester

13. On the toolbar, click Import.

14. In the Data Processing dialog box, once the data has been successfully imported, click Close.

15. Arrange the diagram so that you can see the Ship Date and Reseller Sales tables, and then drag the
ShipDateKey column from the Reseller Sales table to the DateKey column in the Ship Date table to
create the relationship.

 Task 2: Rename and Hide Columns

1. Click the title bar of the Geography table, and then click its Maximize icon.
2. In the maximized Geography table, click the GeographyKey column, hold down CTRL, and then click
the SalesTerritoryKey column to select both columns.

3. Right-click either of the selected columns, and then click Hide from Client Tools.

4. In the maximized Geography table, right-click the StateProvinceCode column, click Rename, and
then rename the column to State or Province Code.

5. Repeat the previous step to rename the following columns in the Geography table:

Column New name

StateProvinceName State or Province

CountryRegionCode Country or Region Code

EnglishCountryRegionName Country or Region

 MCT USE ONLY. STUDENT USE PROHIBITED
L7-8 Developing SQL Data Models

Column New name

PostalCode Postal Code

6. Click the Restore icon for the Geography table.

7. Click the title bar of the Reseller table, and then click its Maximize icon.

8. In the maximized Reseller table, click the ResellerKey column, hold down CTRL, and then click the
GeographyKey column and the ResellerAlternateKey column to select the columns.

9. Right-click any of the selected columns, and then click Hide from Client Tools.
10. In the maximized Reseller table, right-click the BusinessType column, click Rename, and then
rename the column to Business Type.

11. Repeat the previous step to rename the following columns in the Reseller table:

Column New name

ResellerName Reseller Name

NumberEmployees Employees

12. Click the Restore icon for the Reseller table.

13. Click the title bar of the Employee table, and then click its Maximize icon.

14. In the maximized Employee table, click the EmployeeKey column, hold down CTRL, and then click
the ParentEmployeeKey column, the EmployeeNationalIDAlternateKey column, and the
ParentEmployeeNationalIDAlternateKey column to select the columns.

15. Right-click any of the selected columns, and then click Hide from Client Tools.

16. In the maximized Employee table, right-click the FirstName column, click Rename, and then rename
the column to First Name.

17. Repeat the previous step to rename the following columns in the Employee table:

Column New name

LastName Last Name

MiddleName Middle Name

DepartmentName Department

18. Click the Restore icon for the Employee table.

19. Click the title bar of the Order Date table, and then click its Maximize icon.

20. In the maximized Order Date table, click the DateKey column, hold down CTRL, and then click the
DayNumberOfWeek column and the MonthNumberOfYear column to select the columns.

21. Right-click any of the selected columns, and then click Hide from Client Tools.
 MCT USE ONLY. STUDENT USE PROHIBITED
L7-9

22. In the maximized Order Date table, right-click the FullDateAlternateKey column, click Rename, and
then rename the column to Date.

23. Repeat the previous step to rename the following columns in the Order Date table:

Column New name

EnglishDayNameOfWeek Weekday

DayNumberOfMonth Day of Month

EnglishMonthName Month

CalendarQuarter Calendar Quarter

CalendarYear Calendar Year

FiscalQuarter Fiscal Quarter

FiscalYear Fiscal Year

24. Click the Restore icon for the Order Date table.

25. Click the title bar of the Ship Date table, and then click its Maximize icon.
26. In the maximized Ship Date table, click the DateKey column, hold down CTRL, and then click the
DayNumberOfWeek column and the MonthNumberOfYear column to select the columns.

27. Right-click any of the selected columns, and then click Hide from Client Tools.

28. In the maximized Ship Date table, right-click the FullDateAlternateKey column, click Rename, and
then rename the column to Date.

29. Repeat the previous step to rename the following columns in the Ship Date table:

Column New name

EnglishDayNameOfWeek Weekday

DayNumberOfMonth Day of Month

EnglishMonthName Month

CalendarQuarter Calendar Quarter

CalendarYear Calendar Year

FiscalQuarter Fiscal Quarter

FiscalYear Fiscal Year

30. Click the Restore icon for the Ship Date table.

31. Click the title bar of the Product table, and then click its Maximize icon.
 MCT USE ONLY. STUDENT USE PROHIBITED
L7-10 Developing SQL Data Models

32. In the maximized Product table, click the ProductKey column, hold down CTRL, and then click the
ProductAlternateKey column and the ProductSubcategoryKey column to select the columns.

33. Right-click any of the selected columns, and then click Hide from Client Tools.

34. In the maximized Product table, right-click the EnglishProductName column, click Rename, and
then rename the column to Product Name.

35. Repeat the previous step to rename the following columns in the Product table:

Column New name

StandardCost Standard Cost

ListPrice List Price

SizeRange Size Range

EnglishDescription Description

36. Click the Restore icon for the Product table.

37. Click the title bar of the Product Subcategory table, and then click its Maximize icon.
38. In the maximized Product Subcategory table, click the ProductSubcategoryKey column, hold
down CTRL, and then click the ProductSubcategoryAlternateKey column and the
ProductCategoryKey column to select the columns.
39. Right-click any of the selected columns, and then click Hide from Client Tools.

40. In the maximized Product Subcategory table, right-click the EnglishProductSubcategoryName

column, click Rename, and then rename the column to Subcategory.
41. Click the Restore icon for the Product Subcategory table.

42. Click the title bar of the Product Category table, and then click its Maximize icon.
43. In the maximized Product Category table, click the ProductCategoryKey column, hold down CTRL,
and then click the ProductCategoryAlternateKey column to select both columns. Right-click either
of the selected columns, and then click Hide from Client Tools.

44. In the maximized Product Category table, right-click the EnglishProductCategoryName column,
click Rename, and then rename the column to Category.

45. Click the Restore icon for the Product Category table.

46. Click the title bar of the Reseller Sales table, and then click its Maximize icon.

47. In the maximized Reseller Sales table, click the ProductKey column, hold down SHIFT, and then click
the ShipDate column to select all columns other than the Quantity, Cost, and Revenue measures
that you created in the previous exercise.

48. Right-click any of the selected columns, and then click Hide from Client Tools.

49. Click the Restore icon for the Reseller Sales table.

 Task 3: Configure Column Sort Order

1. On the Model menu, point to Model View, and then click Data View.
2. On the Ship Date tab, click the column heading for the Weekday column.
 MCT USE ONLY. STUDENT USE PROHIBITED
L7-11

3. On the Column menu, point to Sort, and then click Sort by Column.

4. In the Sort by Column dialog box, in the Sort column, ensure that Weekday is selected, in the By
column, select DayNumberOfWeek, and then click OK.

5. On the Ship Date tab, click the column heading for the Month column.

6. On the Column menu, point to Sort, and then click Sort by Column.
7. In the Sort by Column dialog box, in the Sort column, ensure that Month is selected, in the By
column, select MonthNumberOfYear, and then click OK.

8. On the Order Date tab, click the column heading for the Weekday column.

9. On the Column menu, point to Sort, and then click Sort by Column.

10. In the Sort by Column dialog box, in the Sort column, ensure that Weekday is selected, in the By
column, select DayNumberOfWeek, and then click OK.

11. On the Order Date tab, click the column heading for the Month column.

12. On the Column menu, point to Sort, and then click Sort by Column.

13. In the Sort by Column dialog box, in the Sort column, ensure that Month is selected, in the By
column, select MonthNumberOfYear, and then click OK.

 Task 4: Create Hierarchies

1. On the Model menu, point to Model View, and then click Diagram View.

2. In the diagram, click the title bar of the Geography table, and then click its Maximize icon.
3. In the title bar of the maximized Geography table, click the Create Hierarchy icon, and then rename
the new hierarchy to Location.
4. In the maximized Geography table, drag the Country or Region column onto the Location
hierarchy, drag the State or Province column onto the Location hierarchy, drag the City column
onto the Location hierarchy, and then drag the Postal Code column onto the Location hierarchy.

5. Click the Restore icon for the Geography table.

6. Click the title bar of the Order Date table, and then click its Maximize icon.

7. In the title bar of the maximized Order Date table, click the Create Hierarchy icon, and then rename
the new hierarchy to Calendar Date.

8. In the maximized Order Date table, drag the following columns onto the Calendar Date hierarchy, in
the same order:

o Calendar Year

o Calendar Quarter

o Month

o Day of Month

9. In the title bar of the maximized Order Date table, click the Create Hierarchy icon, and then rename
the new hierarchy to Fiscal Date.

10. In the maximized Order Date table, drag the following columns onto the Fiscal Date hierarchy, in
the same order:

o Fiscal Year

o Fiscal Quarter
 MCT USE ONLY. STUDENT USE PROHIBITED
L7-12 Developing SQL Data Models

o Month

o Day of Month

11. Click the Restore icon for the Order Date table.

 Task 5: Test the Model

1. On the Model menu, click Analyze in Excel.

2. In the Analyze in Excel dialog box, ensure that Current Windows User is selected and that the
(Default) perspective is specified, and then click OK.

3. In Excel, in the PivotTable Fields pane, select Revenue. The total revenue is shown in the PivotTable.
4. In the Order Date table, select the Calendar Date hierarchy. The revenue for each calendar year is
shown.

5. In the PivotTable, expand 2007 to reveal the quarterly revenue, and then expand quarter 3 to show
the monthly revenue. Note that the months are displayed in chronological order.
6. Close Excel without saving the workbook.

Results: After completing this exercise, you should have a tabular data model that includes renamed
columns, custom sort orders, and specific relationships between tables.

Exercise 3: Deploying a Tabular Data Model Project in SQL Server Analysis

Services
 Task 1: Separate Data into Partitions
1. On the Model menu, point to Model View, and then click Data View.
2. On the Reseller Sales tab, on the Table menu, and then click Partitions.

3. In the Partition Manager dialog box, ensure that the Reseller Sales table is selected, and then click
New.
4. In the Partition Name box, change the text to Since 2008.

5. In the Partition Name column, click Partition, and then in the Partition Name box, change the text
to Before 2008, and then click OK.

6. On the Table menu, and then click Partitions.

7. In the Partition Manager dialog box, ensure that the Before 2008 partition is selected, and then
click the Design button.

8. In the Query Editor, click the drop-down selector in the OrderDateKey cell, click Number Filters,
and then click Less Than.

9. In the Filter Rows dialog box, in the box to the right of is less than type 20080101, then click OK.
10. On the toolbar, click Import.

11. In the Partition Manager dialog box, click the Since 2008 partition is selected, and then click the
Design button.

12. In the Query Editor, click the drop-down selector in the OrderDateKey cell, click Number Filters,
and then click Greater Than.
 MCT USE ONLY. STUDENT USE PROHIBITED
L7-13

13. In the Filter Rows dialog box, in the box to the right of is greater than type 20080101, then click
OK.

14. On the toolbar, click Import.

15. In the Partition Manager dialog box, click OK.

16. On the Model menu, point to Process, and then click Process Partitions.
17. In the Process Partitions dialog box, for Mode, select Process Full, in the Process column, select
Before 2008 and Since 2008, and then click OK. If you are prompted for impersonation credentials,
type the user name ADVENTUREWORKS\ServiceAcct and the password Pa55w.rd, and then click
OK.

18. In the Data Processing dialog box, wait until the processing is complete, and then observe how
many rows were transferred for each partition, and then click Close.

 Task 2: Create a Perspective

1. On the Model menu, point to Perspectives, and then click Create and Manage.

2. In the Perspectives dialog box, click New Perspective, and then type Products Analysis.
3. In the Products Analysis column, select Product, Product Category, Product Subcategory,
Reseller Sales, and then click OK.

4. On the Model menu, click Analyze in Excel.

5. In the Analyze in Excel dialog box, ensure that Current Windows User is selected, for Perspective,
select Products Analysis, and then click OK.

6. In Excel, in the PivotTable Fields pane, note that only the tables that you specified for the
perspective are visible.

7. Under Reseller Sales, select Revenue.

8. In the Product Category table, select Category.

9. Close Excel without saving the workbook.

 Task 3: Create a Security Role

1. On the Model menu, click Roles.

2. In the Role Manager dialog box, click New.

3. In the Name column, type US Bike Sales, and then in the Permissions column, select Read and
Process.

4. In the DAX Filter column, for the Geography table, type the DAX expression as shown below:

='Geography'[Country or Region Code]="US"

5. In the DAX Filter column, for the Product Category table, type the DAX expression as shown below,
and then click OK:

='Product Category'[Category]="Bikes"

6. On the Model menu, click Analyze in Excel.

7. In the Analyze in Excel dialog box, click Role, and then in the drop-down list below Role, select US
Bike Sales, and then click OK.

8. Ensure that the (Default) perspective is specified, and then click OK.
 MCT USE ONLY. STUDENT USE PROHIBITED
L7-14 Developing SQL Data Models

9. In Excel, in the PivotTable Fields pane, under Reseller Sales, select Revenue.

10. In the Geography table, select Location.

11. In the Product Category table, select Category.

12. In the Product Subcategory table, select Subcategory.

13. Note that reseller data is only showing where Location is United States and Category is Bikes.

14. Close Excel without saving the workbook.

 Task 4: Deploy the Reseller Sales Project

1. In SQL Server Data Tools, in Solution Explorer, right-click the AWSalesTab project, and then click
Properties.

2. In the AWSalesTab Property Pages dialog box, in the Deployment Server section, verify that the
Server value is localhost\SQL2.

3. Change the Database property to AdventureWorksTab, change the Model Name property to
Reseller Sales, and then click OK.

4. On the Build menu, click Deploy AWSalesTab. If you are prompted for impersonation credentials,
type the user name ADVENTUREWORKS\ServiceAcct and the password Pa55w.rd, and then click
OK.

5. In the Deploy dialog box, when deployment has completed, click Close.

6. Close SQL Server Data Tools.

 Task 5: Use the Deployed Tabular Database

1. Start Excel, and then create a new blank workbook.

2. On the Data tab, in the Get External Data area, in the From Other Sources drop-down list, select
From Analysis Services.
3. In the Data Connection Wizard dialog box, in the Server name box, type MIA-SQL\SQL2, ensure
that Use Windows Authentication is selected, and then click Next.
4. On the Select Database and Table page, ensure that the AdventureWorksTab database and the
Reseller Sales cube are selected, and then click Next.

5. On the Save Data Connection File and Finish page, click Finish.

6. In the Import Data dialog box, ensure that the Existing worksheet option is selected, and then click
OK.

7. In the PivotTable Fields pane, under Reseller Sales, select Revenue.

8. In the Geography table, select the Location hierarchy.

9. In the Order Date table, drag the Fiscal Date hierarchy to the Columns area of the PivotTable Fields
pane.

10. Explore the data in the PivotTable. When you have finished, close Excel without saving the workbook.

Results: After this exercise, you should have deployed the tabular data model project.
 MCT USE ONLY. STUDENT USE PROHIBITED
L8-1

Module 8: Introduction to Data Analysis Expressions (DAX)

Lab: Using DAX to Enhance a Tabular Data
Model
Exercise 1: Creating Calculated Columns
 Task 1: Prepare the Lab Environment
1. Ensure the 20768C-MIA-DC and 20768C-MIA-SQL virtual machines are both running, and then log
on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd.

2. In the D:\Labfiles\Lab08\Starter folder, right-click Setup.cmd and click Run as administrator.

3. In the User Account Control dialog box, click Yes, and then wait for the script to finish.

 Task 2: Concatenate Text Values

1. Start Visual Studio 2015 and open the AWSalesTab.sln solution in the D:\Labfiles\Lab08\Starter
folder.

2. If the Tabular model designer dialog box is displayed, in the Workspace server list, select
localhost\SQL2, and then click OK.
3. If the Tabular Model Designer dialog box appears, prompting you to run a script on the server, click
Yes.

4. In Solution Explorer, double-click Model.bim to open the data model designer.

5. On the Model menu, point to Process, and click Process All. If the Impersonation Credentials
dialog box appears, use the following credentials, and then click OK:

o User name: ADVENTUREWORKS\ServiceAcct

o Password: Pa55w.rd

6. In the Data Processing dialog box, when all of the tables have been processed, click Close.

7. On the Employee tab, after the existing columns, double-click the Add Column header, type
Employee Name, and then press Enter.

8. Click the Employee Name column, click the formula bar, and type the following DAX expression, and
then press Enter:

=[First Name] & IF(ISBLANK([Middle Name]), "", CONCATENATE(" ", [Middle Name])) &
CONCATENATE(" ", [Last Name])

Verify that the new column shows the full employee name (including middle name, if present).
9. Click the First Name column header, and then hold Shift and click the Middle Name column header
to select the First Name, Last Name, and Middle Name columns. Then right-click any of the
selected columns and click Hide from Client Tools.

10. On the File menu, click Save All.

 MCT USE ONLY. STUDENT USE PROHIBITED
L8-2 Developing SQL Data Models

 Task 3: Calculate a Numeric Value

1. On the Reseller Sales tab, in the first empty column to the right of the populated columns, double-
click Add Column, type SalesProfit, and then press Enter.

2. Click the SalesProfit column, then click the formula bar, and type the following DAX expression, and
then press Enter:

=[SalesAmount]-[TotalProductCost]

Wait for the table to finish updating, and note the calculated values in the new column.

3. Press F4 and view the Properties pane to verify that the Data Format property for this column has
been set to Currency.

 Task 4: Show Related Values

1. On the Product tab, after the existing columns, double-click Add Column, type Subcategory, and
then press Enter.

2. Click Subcategory, click the formula bar and type the following DAX expression, and then press
Enter:

=RELATED('Product Subcategory'[Subcategory])

Scroll down to verify that products with a subcategory are shown with the subcategory name. Note
that some products at the beginning of the table are not members of a subcategory, so have a blank
value in the new column.
3. After the Subcategory column, double-click Add Column, type Category, then press Enter.

4. Click Category, click the formula bar, and type the following DAX expression, and then press Enter:

=RELATED('Product Category'[Category])

Scroll down to verify that products with a category are shown with the category name. Note that
some products at the beginning of the table are not members of a category, so have a blank value in
the new column.

5. On the Model menu, point to Model View, and then click Diagram View.
6. In the diagram, click the title bar of the Product table, then click its Maximize icon.

7. On the title bar of the maximized Product table, click the Create Hierarchy icon, type Categorized
Products, and then press Enter.
8. Drag the following columns from the column list of the Product table to the Categorized Products
hierarchy:

o Category

o Subcategory

o Product Name

9. On the Product table title bar, click the Restore icon.

10. On the Model menu, point to Model View, and then click Data View.

11. On the Product Subcategory tab, right-click the Subcategory column—which is the only visible
column—and click Hide from Client Tools.
 MCT USE ONLY. STUDENT USE PROHIBITED
L8-3

12. In the Product Category table, right-click the Category column—which is the only visible column—
and click Hide from Client Tools.
13. On the File menu, click Save All.

 Task 5: View Calculated Columns in Excel

1. On the Model menu, click Analyze in Excel.

2. In the Analyze in Excel dialog box, ensure that Current Windows User is selected and that the
(Default) perspective is specified, and click OK.

3. In Excel, in the PivotTable Fields pane, under Reseller Sales, select the Revenue measure.
4. Under Product, select Categorized Products so that the PivotTable shows sales revenue for each
product category.

5. Under Employee, drag the Employee Name to the Columns area so that the PivotTable shows sales
revenue for each employee.
6. Expand product categories and subcategories to verify that the hierarchy shows subcategory and
product names.

7. Close Excel without saving the workbook.

8. Leave Visual Studio open for the next exercise.

Results: After this exercise, you should have a calculated column named Employee Name in the
Employee table, a calculated measure named SalesProfit in the Reseller Sales table, and a hierarchy
named Categorized Products in the Product table.

Exercise 2: Creating Measures

 Task 1: Create a Measure That Aggregates a Column
1. In Visual Studio, in the data model designer, on the Reseller Sales tab, click the first empty cell in the
measure grid under the SalesProfit column.

2. In the formula bar, type the following DAX expression, and then press Enter:

Profit:=SUM([SalesProfit])

3. Select the cell containing the Profit measure. In the Properties pane, set the Format property to
Currency.
4. Right-click the column header for the SalesProfit column and click Hide from Client Tools.

5. On the File menu, click Save All.

 Task 2: Create a Measure That References Other Measures

1. On the Reseller Sales tab, click the empty cell in the measure grid under the Profit measure.
2. In the formula bar, type the following DAX expression, and then press Enter:

Margin:=[Profit]/[Revenue]
 MCT USE ONLY. STUDENT USE PROHIBITED
L8-4 Developing SQL Data Models

3. Select the cell containing the Margin measure, and in the Properties pane, set Format to
Percentage.
4. On the File menu, click Save All.

 Task 3: Create a Measure That Uses Time Intelligence

1. In the data model designer, click the Order Date tab.

2. On the Table menu, point to Date, and then click Mark As Date Table.

3. In the Mark as Date Table dialog box, in the Date list, ensure is Date selected, and then click OK.

4. On the Reseller Sales tab, click the empty cell in the measure grid under the Revenue measure.

5. In the formula bar, type the following DAX expression, and then press Enter:

Previous Year Revenue:=CALCULATE([Revenue], SAMEPERIODLASTYEAR('Order Date'[Date]))

6. Select the cell containing the Previous Year Revenue measure, and in the Properties pane, set the
Format property to Currency.

7. On the File menu, click Save All.

 Task 4: View Measures in Excel

1. On the Model menu, click Analyze in Excel.

2. In the Analyze in Excel dialog box, ensure that Current Windows User is selected and that the
(Default) perspective is specified, and click OK.
3. In Excel, in the PivotTable Fields section, under Reseller Sales table, select the Revenue, Profit,
Margin, and Previous Year Revenue measures.
4. Under Order Date, drag the Calendar Date hierarchy to the Rows area so that the PivotTable shows
revenue, profit, margin, and previous year revenue for each year.

5. Expand 2006 and 2007, and note that the revenue for each quarter in 2006 is shown as the previous
year revenue for the same quarters in 2007.

6. Close Excel without saving the workbook. Leave Visual Studio open for the next exercise.

Results: At the end of this exercise, the Reseller Sales table should contain the following measures:

Profit

Margin

Previous Year Revenue

 MCT USE ONLY. STUDENT USE PROHIBITED
L8-5

Exercise 3: Creating a KPI

 Task 1: Create a Measure to Calculate a KPI Goal
1. In Visual Studio, in the data model designer, on the Reseller Sales tab, click the empty cell in the
measure grid under the Previous Year Revenue measure.

2. In the formula bar, type the following DAX expression, and then press Enter:

Revenue Goal:=[Previous Year Revenue] * 1.2

3. Select the cell containing the Revenue Goal measure, and in the Properties pane, set the Format
property to Currency.

4. Right-click the cell containing the Revenue Goal measure, and click Hide from Client Tools.

5. On the File menu, click Save All.

 Task 2: Create a KPI

1. On the Reseller Sales tab, right-click the cell in the measure grid containing the Revenue measure,
and click Create KPI.

2. In the Key Performance Indicator (KPI) dialog box, note that the KPI base measure (value) is
defined by the Revenue measure. Under Target, ensure that Measure is selected, and in the list,
select the Revenue Goal measure.
3. On the red, amber, and green bar, drag the first status threshold slider to 75% and the second to
95%. Note the default icon style, and click OK. Notice that the Revenue measure gains an icon in the
measure grid to indicate that it is the basis for a KPI.
4. On the File menu, click Save All.

 Task 3: View a KPI in Excel

1. On the Model menu, click Analyze in Excel.
2. In the Analyze in Excel dialog box, ensure that Current Windows User is selected and that the
(Default) perspective is specified, and click OK.

3. In Excel, in the PivotTable Fields pane, expand KPIs, expand Revenue, and then select the Value,
Goal, and Status check boxes.

4. Under Order Date, drag the Fiscal Date hierarchy to the Rows area so that the PivotTable shows the
KPI data for each year.

5. Expand 2008 and its quarters and months to view the status of the KPI for each month.

6. Close Excel without saving the workbook. Leave Visual Studio open for the next exercise.

Results: At the end of this exercise, the Reseller Sales table should include a measure named Revenue
Goal and a KPI based on the Revenue Goal measure.
 MCT USE ONLY. STUDENT USE PROHIBITED
L8-6 Developing SQL Data Models

Exercise 4: Implementing a Parent-Child Hierarchy

 Task 1: Create a Path Column
1. In Visual Studio, in the data model designer, click the Employee tab.

2. After the existing columns, double-click Add Column, type Path, and then press Enter.

3. Click Path, click the formula bar, type the following DAX expression, and then press Enter:

=PATH([EmployeeKey], [ParentEmployeeKey])

Wait for the table to finish updating, and note the calculated values in the new column.

4. On the File menu, click Save All.

 Task 2: Create a Column for Each Hierarchy Level

1. In the Employee table, in the first empty column to the right of the populated columns, double-click
Add Column, type Level1, then press Enter.

2. Click Level1, click the formula bar, type the following DAX expression, and then press Enter:

=LOOKUPVALUE ([Employee Name], [EmployeeKey], PATHITEM ([Path], 1, 1))

3. In the first empty column to the right, double-click Add Column, type Level2, then press Enter.
4. Click Level2, click the formula bar, type the following DAX expression, and then press Enter:

=LOOKUPVALUE ([Employee Name], [EmployeeKey], PATHITEM ([Path], 2, 1))

5. In the first empty column to the right, double-click Add Column, type Level3, then press Enter.

6. Click Level3, click the formula bar, type the following DAX expression, and then press Enter:

=LOOKUPVALUE ([Employee Name], [EmployeeKey], PATHITEM ([Path], 3, 1))

7. In the first empty column to the right, double-click Add Column, type Level4, then press Enter.

8. Click Level4, click the formula bar, type the following DAX expression, and then press Enter:

=LOOKUPVALUE ([Employee Name], [EmployeeKey], PATHITEM ([Path], 4, 1))

9. On the File menu, click Save All.

 Task 3: Use the Calculated Columns in a Hierarchy

1. On the Model menu, point to Model View, and then click Diagram View.

2. Click the Employee table, and then click the Maximize button.

3. In the maximized Employee table, click Create Hierarchy, type Employee Hierarchy, and then press
Enter.

4. Drag the following columns from the column list of the Employee table to the Employee Hierarchy,
in the following order:

a. Level1

b. Level2

c. Level3

d. Level4
 MCT USE ONLY. STUDENT USE PROHIBITED
L8-7

5. Click the Path column, hold Shift and click the Level4 column. Then right-click the selected columns
and click Hide from Client Tools.
6. In the title bar, click Restore.

7. On the File menu, click Save All.

 Task 4: View a Parent-Child Hierarchy in Excel

1. On the Model menu, click Analyze in Excel.

2. In the Analyze in Excel dialog box, ensure that Current Windows User is selected and that the
(Default) perspective is specified, and click OK.

3. In Excel, in the PivotTable Fields section, under Reseller Sales, select the Revenue measure.
4. Under Employee, select Employee Hierarchy so that the PivotTable shows revenue for the top-level
employee.
5. Expand the employee hierarchy to view sales totals for managers and their subordinates. Note that
the personal revenue for sales managers is shown with a blank employee name under the total for
that sales manager.

6. Close Excel without saving the workbook.

7. Close Visual Studio.

Results: At the end of this exercise, the Employee table should include a hierarchy named Employee
Hierarchy.
MCT USE ONLY. STUDENT USE PROHIBITED
 
 MCT USE ONLY. STUDENT USE PROHIBITED
L9-1

Module 9: Performing Predictive Analysis with Data Mining

Lab: Using Data Mining to Support a
Marketing Campaign
Exercise 1: Creating a Data Mining Structure and Models
 Task 1: Prepare the Lab Environment
1. Ensure that the 20768C-MIA-DC, and 20768C-MIA-SQL virtual machines are running.

2. Log on to 20768C-MIA-SQL as ADVENTUREWORKS\Student with the password Pa55w.rd.

3. In the D:\Labfiles\Lab09\Starter folder, right-click Setup.cmd, and then click Run as

administrator.

4. In the User Account Control dialog box, click Yes, and then wait for the script to finish.

 Task 2: Create a Data Mining Project

1. Start Visual Studio 2015, and on the File menu, point to New, and click Project.

2. In the New Project dialog box, expand Business Intelligence, then click Analysis Services, then
click Analysis Services Multidimensional and Data Mining Project.
3. Type AW Data Mining in the Name box, then type D:\Labfiles\Lab09\Starter in the Location box,
and then click OK.

4. In Solution Explorer, right-click Data Sources, and then click New Data Source.
5. In the Data Source Wizard, on the Welcome to the Data Source Wizard page, click Next.

6. On the Select how to define the connection page, click Create a data source based on an
existing or new connection, and then click New.

7. In the Connection Manager dialog box, in the Server name field, type MIA-SQL, in the Select or
enter a database name drop-down list, select AdventureWorksDW, and click OK.

8. In the Data Source Wizard, on the Select how to define the connection page, click Next.

9. On the Impersonation Information page, click Use the service account, and then click Next.

10. On the Completing the Wizard page, click Finish.

11. In Solution Explorer, right-click Data Source Views, and then click New Data Source View.

12. In the Data Source View Wizard, on the Welcome to the Data Source View Wizard page, click
Next.

13. On the Select a Data Source page, ensure that Adventure Works DW is selected, and then click
Next.

14. On the Select Tables and Views page, in the Available objects list, click ProspectiveBuyer (dbo),
hold the Ctrl key. Click vTargetMail (dbo) and click the > button to move the selected objects to the
Included objects list, and then click Next.

15. On the Completing the Wizard page, in the Name field, type Adventure Works DW DM View,
and then click Finish.
 MCT USE ONLY. STUDENT USE PROHIBITED
L9-2 Developing SQL Data Models

 Task 3: Create a Data Mining Structure and a Data Mining Model

1. In Solution Explorer, right-click Mining Structures, click New Mining Structure.

2. In the Data Mining Wizard, on the Welcome to the Data Mining Wizard page, click Next.

3. On the Select the Definition Method page, ensure that From existing relational database or data
warehouse is selected, and then click Next.
4. On the Create the Data Mining Structure page, ensure that Create mining structure with mining
model is selected and select the Microsoft Decision Trees data mining technique. Then click Next.

5. On the Select a Data Source View page, select Adventure Works DW DM View and click Next.
6. On the Specify Table Types page, in the vTargetMail row, select the check box in the Case column,
and then click Next.

7. On the Specify the Training Data page, in the Mining model structure table, in the BikeBuyer
row, select the check box in the Predictable column.
8. In the CustomerKey row, select the check box in the Key column; for all other rows, select the Input
check box, and then click Next.
9. On the Specify Columns’ Content and Data Type page, click Detect. Ensure that the content type
of the Bike Buyer column is identified as Discrete, and change the Yearly Income column to
Discrete. Then click Next.
10. On the Create Testing Set page, note that the Percentage of data for testing value is 30 percent,
and then click Next.

11. On the Completing the Wizard page, in the Mining structure name field, type Purchase
Prediction.
12. In the Mining model name field, type Purchase Decision Tree, and then click Finish.

13. On the Build menu, click Deploy AW Data Mining.

14. When deployment is complete, close the Deployment Progress - AW Data Mining window.

 Task 4: Add a Data Mining Model to a Data Mining Structure

1. In Solution Explorer, expand Mining Structures and double-click Purchase Prediction.dmm.

2. In the Purchase Prediction.dmm [Design] window, on the Mining Models tab, right-click
anywhere, and then click New Mining Model.

3. In the New Mining Model dialog box, in the Model name box, type Purchase - Bayes. Select
Microsoft Naive Bayes in the Algorithm name box, and then click OK.

4. In the Microsoft Visual Studio dialog box, read the information message about content types
unsupported by the Microsoft Naïve Bayes algorithm, then click Yes.

5. In the Mining Models tab, in the Purchase - Bayes column, in the Name Style row, select Ignore to
exclude the column from the model.

6. On the Build menu, click Deploy AW Data Mining.

7. When deployment is complete, close the Deployment Progress - AW Data Mining window.

8. Leave Visual Studio open for the next exercise.

Results: After this exercise, you should have a data mining structure with two data models.
 MCT USE ONLY. STUDENT USE PROHIBITED
L9-3

Exercise 2: Explore a Data Mining Model

 Task 1: Review Data Models
1. In Solution Explorer, expand Mining Structures, right-click Purchase Prediction.dmm, and then
click Browse.

2. In the Mining Model drop-down list, ensure that Purchase Decision Tree is selected, and on the
Dependency Network tab, move the slider gradually from All Links to Strongest Links to see which
factors are the strongest predictors that a customer will purchase a bike. You might need to move the
Mining Legend window to be able to see the Dependency Network tab.
3. On the Decision Tree tab, in the Background drop-down list, select 1. Then view the decision tree to
see how the other column values influence a value of 1 for Bike Buyer.
4. In the Mining Model drop-down list, select Purchase - Bayes, and on the Attribute Profiles tab,
view the color-coded indicators of the values for each column when compared to customers with a
Bike Buyer value of 1 or 0.
5. On the Attribute Characteristics tab, in the Attribute drop-down list, ensure that Bike Buyer is
selected, and in the Value drop-down list, select 1. Then view the probability for each of the other
column values when the Bike Buyer value is 1.
6. On the Attribute Discrimination tab, in the Attribute drop-down list, ensure that Bike Buyer is
selected. In the Value 1 drop-down list, select 1, and in the Value 2 drop-down list, select 0. Then
note how values for all other columns favor a particular Bike Buyer value.

7. Leave Visual Studio open for the next exercise.

Results: After this exercise, you should have reviewed the output of data mining models using Visual
Studio.

Exercise 3: Validating Data Mining Models

 Task 1: Review a Lift Chart
1. In Visual Studio, in the Purchase Prediction.dmm [Design] pane, click Mining Accuracy Chart.
2. On the Input Selection tab, verify that the Show box is selected for both the Purchase Decision
Tree and Purchase – Bayes models, and that Use mining model test cases is selected in the Select
data set to be used for Accuracy Chart section.

3. On the Lift Chart tab, review the lift chart, and the detailed data in the Mining Legend window.

The Purchase Decision Tree model is closer to the Ideal Model, and can therefore be considered
more accurate.

 Task 2: Review a Profit Chart

1. On the Lift Chart tab, in the Chart type drop-down list, select Profit Chart. You might need to move
the Mining Legend window to see the Chart type list.
2. In the Profit Chart Settings dialog box, enter the following values to reflect a marketing campaign
you are planning, and then click OK:

o Population: 5,000 (this is the number of potential customers you plan to contact).
 MCT USE ONLY. STUDENT USE PROHIBITED
L9-4 Developing SQL Data Models

o Fixed cost: 500 (this is the fixed cost of your marketing campaign).

o Individual cost: 1 (this is the cost associated with contacting each customer).

o Revenue per individual: 150 (this is the amount you expect a customer to spend if they respond
positively to the campaign).

3. Review the chart and the Mining Legend window to evaluate which mining model is likely to
generate the most profitable marketing campaign based on the test data.

 Task 3: Review a Classification Matrix

1. Click the Classification Matrix tab.

2. Review the matrix, noting that for each model it shows the number of times the model predicted a
Bike Buyer value of 1 or 0 on rows, with columns for the actual value of the Bike Buyer column in
the test data.

 Task 4: Review a Cross-Validation Report

1. Click the Cross Validation tab.
2. Enter the following values, and click Get Results:

o Fold Count: 5 (this is the number of partitions used to group the data for analysis).

o Max Cases: 500 (this is the number of cases to be analyzed).

o Target Attribute: Bike Buyer (this is the predictable column to be evaluated).

o Target State: 1 (this is the desired value for the target attribute).
o Target Threshold: 0.7 (this is a value between 0 and 1 that indicates the level of accuracy
required for a prediction to be considered correct).

3. View the resulting report, and note that for each mining model, the results include the following:

o Classifications for true positives, false positives, true negatives, and false negatives.

o The likely lift gained by using the model.

4. Leave Visual Studio open for the next exercise.

Results: After this exercise, you should have two validated data mining models in your data mining
structure.

Exercise 4: Using a Data Mining Model in a Report

 Task 1: Create a Report
1. In Visual Studio, on the File menu, point to New and click Project.
2. In the New project dialog box, in the Templates pane, expand Business Intelligence, then select
Report Server Project Wizard.

3. In the Name field, type Promotion Targeting, in the Location field, type
D:\Labfiles\Lab09\Starter, and then click OK.

4. In the Report Wizard, on the Welcome to the Report Wizard page, click Next.
 MCT USE ONLY. STUDENT USE PROHIBITED
L9-5

5. On the Select the Data Source page, select New data source, change the Name to
AWDataMining.
6. In the Type drop-down list, select Microsoft SQL Server Analysis Services, and click Edit.

7. In the Connection Properties dialog box, in the Server name field, type MIA-SQL, in the Select or
enter a database name drop-down list, click AW Data Mining, and then click OK.

8. On the Select the Data Source page, click Next.

9. On the Design the Query page, click Query Builder.

10. In Query Designer, in the Mining Model pane, click Select Model.

11. In the Select Mining Model dialog box, expand Purchase Prediction, click Purchase - Bayes, and
click OK.

12. In the Select Input Table(s) pane, click Select Case Table.

13. In the Select Table dialog box, click ProspectiveBuyer (dbo), and click OK.

14. In the table at the bottom of the query designer, in the Source column, select ProspectiveBuyer
table, and then in the Field column, in the drop-down list, click FirstName.

15. Repeat step 14 to add five more rows to the table by using the settings in the following table, and
then click OK:

Source Field Alias Criteria/Argument

ProspectiveBuyer table LastName

ProspectiveBuyer table Address Line 1

ProspectiveBuyer table City

Purchase – Bayes mining model Bike Buyer =1

Prediction Function PredictProbability Purchase [Purchase -

Probability Bayes].[Bike Buyer]

16. On the Design the Query page, click Next.

17. On the Select the Report Type page, select Tabular, and click Next.

18. On the Design the Table page, move all the columns into the Details area, and click Next.
19. On the Completing the Wizard page, change the Report name to Potential Bike Buyers and click
Finish.

20. In the Row Groups pane, click the (table1_Details_Group) drop-down list and click Group
Properties.

21. In the Group Properties dialog box, on the Sorting tab, click Add, in the Sort by drop-down list,
click [Purchase_Probability], and in the Order drop-down list, click Z to A. Then click OK.

22. In the report design pane, in the Purchase Probability column, right-click [Purchase Probability]
and click Text Box Properties.

23. In the Text Box Properties dialog box, click Number, select Percentage, and click OK.
 MCT USE ONLY. STUDENT USE PROHIBITED
L9-6 Developing SQL Data Models

24. Click the Preview tab to review the Potential Bike Buyers report.

25. When you have finished working on the report, close Visual Studio.

Results: After this exercise, you should have an SSRS report that predicts bike purchasers.