Data Architectures

Pages Introduction

Introduction
The following pages outline several data architectures that can be
deployed as part of QlikView solutions. These pages do not depict
Stage 1 every possible architecture. In fact, they may only represent a small
portion of the possible architectures. They do, however, represent
Stage 2
the concepts that are most often employed (with the exception of the
Stage 1 Architecture) to promote scalability, re-use and consistency.

Stage 3 These pages are meant to help provide a backdrop for design and
architecture discussions related to enterprise deployments with
Stage 4 QlikView.

© 2010 QlikTech International AB. All rights reserved. QlikTech, QlikView, Qlik, Q, Simplifying Analysis for Everyone, Power of Simplicity, New Rules, The Uncontrollable Smile and other QlikTech products and services as well as their respective logos are
trademarks or registered trademarks of QlikTech International AB. All other company names, products and services used herein are trademarks or registered trademarks of their respective owners. The information published herein is subject to change without notice.
This publication is for informational purposes only, without representation or warranty of any kind, and QlikTech shall not be liable for errors or omissions with respect to this publication. The only warranties for QlikTech products and services are those that are set
forth in the express warranty statements accompanying such products and services, if any. Nothing herein should be construed as constituting any additional warranty.
Author: BPN, QlikTech NA
 Data Architectures
Stage 1 Architecture

Source Layer
Description:
This architecture involves only Databases and
direct queries against source other data sources
databases. All presentation
layer applications make
database connections and
contain their own SQL queries
and QlikView scripts to load,
transform and aggregate any
data needed.

When to use:
This is the simplest QlikView
architecture, but also the most
costly to maintain. Queries
Presentation Layer

are often repeated across Production applications. Each one is

many applications. Reloads of scripted with source database queries
applications may be and any transformational scripting
Dashboard Dashboard
competing for the same needed for the final interface. Many
database resources. Intra- elements and metrics overlap across
day reloads will need access applications, requiring redundant code Dashboard
to source databases to load and processing.
from.
Analysis App
Analysis App
PLEASE NOTE:
This architecture is not a recommended best practice for enterprise QlikView
deployments.
© 2010 QlikTech International AB. All rights reserved. QlikTech, QlikView, Qlik, Q, Simplifying Analysis for Everyone, Power of Simplicity, New Rules, The Uncontrollable Smile and other QlikTech products and services as well as their respective logos are
trademarks or registered trademarks of QlikTech International AB. All other company names, products and services used herein are trademarks or registered trademarks of their respective owners. The information published herein is subject to change without notice.
This publication is for informational purposes only, without representation or warranty of any kind, and QlikTech shall not be liable for errors or omissions with respect to this publication. The only warranties for QlikTech products and services are those that are set
forth in the express warranty statements accompanying such products and services, if any. Nothing herein should be construed as constituting any additional warranty.
Author: BPN, QlikTech NA
 Data Architectures
Stage 2 Architecture

Source Layer
Description:
This architecture includes the Databases and other
use of QVD files for a 2nd data data sources
tier. The presentation layer
applications will still need to
generate data models from

Extract Layer
several QVD files, but they
will not need to extract directly QlikView applications that extract
from source databases. This and (optionally) denormalize source
helps promote re-use and tables into QlikView QVD files.
consistency across QVD Generator QVD Generator QVD Generator QVD Generator
presentation layer
applications.

When to use:
QVD Layer

QVD Files – QlikView data file layer.

Use this architecture when
QVDs can be one‐to‐one match with
you want to shield
source tables or denormalized “views”
presentation layer developers
of several source tables QVD Files QVD Files QVD Files QVD Files
from direct database retrievals
and promote re-use. Some
data modeling and scripting
expertise will be needed at
Presentation Layer

Production applications built from

the presentation layer, but not
QVDs in the layer above. No direct
necessarily any source
database queries are needed in these
database query knowledge or
applications and re‐use of common
skills.
QVDs is promoted. Some
understanding of data model best
practices is still required to optimize Analysis App Analysis App
Dashboard
application performance. Dashboard

© 2010 QlikTech International AB. All rights reserved. QlikTech, QlikView, Qlik, Q, Simplifying Analysis for Everyone, Power of Simplicity, New Rules, The Uncontrollable Smile and other QlikTech products and services as well as their respective logos are
trademarks or registered trademarks of QlikTech International AB. All other company names, products and services used herein are trademarks or registered trademarks of their respective owners. The information published herein is subject to change without notice.
This publication is for informational purposes only, without representation or warranty of any kind, and QlikTech shall not be liable for errors or omissions with respect to this publication. The only warranties for QlikTech products and services are those that are set
forth in the express warranty statements accompanying such products and services, if any. Nothing herein should be construed as constituting any additional warranty.
Author: BPN, QlikTech NA
 Data Architectures
Stage 3 Architecture

Source
Layer
Description: Databases and
This architecture includes the other data sources
use of QVD files for the 2nd
and 3rd tiers. The first QVD
layer can be very normalized,

Extract
QlikView applications that extract and

Layer
in some cases it can be a (optionally) denormalize source tables into
one-to-one match of source QlikView QVD files.
DB tables. This layer QVD Generator QVD Generator QVD Generator
accommodates very fast
Base QVD

extractions with minimal

Layer
QVD Files – QlikView data file layer. QVDs can
transformations and
be one‐to‐one match with source tables or
aggregations. The second
denormalized “views” of several source tables Base QVD Base QVD
QVD layer is where
Files Files
aggregated and denormalized
Presentation Transformed Transformation

QVDs are exposed to

presentation layer developers. These QlikView apps read the base QVDs and
perform aggregation and/or denormalization,
Layer

This helps support co-

development of QlikView writing out the final QVDs to be used by
QVD Generator QVD Generator
between IT and business presentation layer applications.
teams.
These QVDs are aggregated, regionalized,
QVD Layer

When to use:
reduced or denormalized to optimize re‐use,
Use this architecture when a
ease of deployment and performance. Denormalized Aggregated Segmented Reduced
short batch window is desired QVDs QVDs QVDs
These are the QVDs that are exposed to QVDs
(QVD layer is fast) and you
presentation layer developers.
want to create the
presentation layer data
models from a QVD layer, but Production applications built from QVDs in
Layer

you want this layer to be the layer above. Some understanding of

optimized for re-use and Analysis Dashboard
data model best practices is still required
performance. to optimize application performance. Dashboard Analysis

Dashboard
 Data Architectures
Stage 4 Architecture

Source
Layer
Description: Databases and other data sources
This architecture includes the
use of “QlikMart” applications.
These QlikMarts are just QlikView applications that extract and

Extract
Layer
QlikView applications without a (optionally) denormalize source tables into
completed user interface. QlikView QVD files.
They act as data models that
QVD Generator QVD Generator QVD Generator
can be loaded into a
presentation layer application
Rich QVD

with a single line of code (called

Layer

a binary load). This helps QVD Files – combination of base QVD files,
support co-development of aggregated, segmented and denormalized QVD
QlikView between IT and files. Aggregated Segmented Base QVD
business teams. QVDs QVDs Files

When to use:
This architecture should be A “QlikMart” is a QlikView application that
used when “self-service BI” is a simply holds a data model that can be
QlikMart
Layer

goal, or when many binary loaded as the base data model for
dashboards will overlap in data end‐user applications. Examples might be QlikMart QlikMart
usage and IT wants to control a Finance QlikMart, a Sales QlikMart and Finance Inventory
QlikMart
the consistency and an Inventory QlikMart. Sales
predictability of dashboard
performance. This architecture
Presentation

shields the presentation layer

Production applications. Built from
designers from data model
Layer

QlikMart applications where possible.

preparation, except in cases
Minimal (if any) scripting. Can optionally
where a QlikMart has not yet
retrieve from QVD layer as well as Analysis
been built. Analysis
QlikMart layer.
Dashboard Dashboard

Dashboard