Pure Storage® FlashArray//m20 Microsoft® SQL Server

Data Warehouse Fast Track Reference Architecture

November 2015
Contents
Executive Summary .................................................................................................................... 5

Goals and Objectives .................................................................................................................. 5

Audience ..................................................................................................................................... 5

Pure Storage Introduction ........................................................................................................... 5

FlashArray//m Specifications.................................................................................................................................. 7

Purity Operating Environment ...............................................................................................................................8

Pure1 ........................................................................................................................................................................8

Experience Evergreen Storage ..............................................................................................................................9

About the SQL Server Data Warehouse Fast Track Program ..................................................... 9

About the Solution ....................................................................................................................... 9

Windows Server Configuration Settings .................................................................................... 10

Storage Layout .......................................................................................................................... 10

Data Reduction on Compressed Databases ............................................................................. 10

SQL Server Configuration Settings ........................................................................................... 12

General Settings and Trace Flags..........................................................................................................................12

Resource Governor Configuration........................................................................................................................12

Database-Specific Settings ....................................................................................................................................12

Solution Hardware & Software Overview .................................................................................. 13

Hardware............................................................................................................................................................... 13

Software................................................................................................................................................................ 13

Bill Of Materials For Cisco C240 M4 Server ......................................................................................................... 13

Certification ............................................................................................................................... 14

Summary ................................................................................................................................... 16

References ................................................................................................................................ 16

© Pure Storage 2015 | 2

About the Author ....................................................................................................................... 17

© Pure Storage 2015 | 3

© 2015 Pure Storage, Inc. All rights reserved. Pure Storage, the "P" Logo, and Pure1 are trademarks or
registered trademarks of Pure Storage, Inc. in the U.S. and other countries. Microsoft, Windows, Windows
Server and SQL Server are registered trademarks of Microsoft Corporation in the U.S. and other countries.
The Pure Storage product described in this documentation is distributed under a license agreement and may
be used only in accordance with the terms of the agreement. The license agreement restricts its use, copying,
distribution, decompilation, and reverse engineering. No part of this documentation may be reproduced in
any form by any means without prior written authorization from Pure Storage, Inc. and its licensors, if any.

THE DOCUMENTATION IS PROVIDED "AS IS" AND ALL EXPRESS OR IMPLIED

CONDITIONS, REPRESENTATIONS AND WARRANTIES, INCLUDING ANY IMPLIED WARRANTY OF
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT ARE DISCLAIMED,
EXCEPT TO THE EXTENT THAT SUCH DISCLAIMERS ARE HELD TO BE LEGALLY INVALID. PURE STORAGE
SHALL NOT BE LIABLE FOR INCIDENTAL OR CONSEQUENTIAL DAMAGES IN CONNECTION WITH THE
FURNISHING, PERFORMANCE, OR USE OF THIS DOCUMENTATION. THE INFORMATION CONTAINED IN THIS
DOCUMENTATION IS SUBJECT TO CHANGE WITHOUT NOTICE.

Pure Storage, Inc. 650 Castro Street, Mountain View, CA 94041

http://www.purestorage.com

© Pure Storage 2015 | 4

Executive Summary
In this document we present a reference architecture for deploying scalable, high-performance Microsoft
SQL Server data warehouses on a Pure Storage FlashArray. Throughout the document we highlight key
tenets of simplicity, ease-of-use, efficiency and reliability of our storage platform.

Our Pure Storage FlashArray//m, on its cost-effective, entry-level model -- the //m20 -- provides the storage
prowess for this solution. On the compute side, we leverage an off-the-shelf, 2 rack-unit, dual CPU socket
Cisco C240 M4 server, running SQL Server 2014 Enterprise Edition Service Pack 1 on top of Windows Server
2012 R2 Standard Edition.

Goals and Objectives

With this document we aim to detail how we have implemented an extremely cost-effective SQL Server
based data warehouses leveraging Pure Storage’s FlashArray//m. We also explain why Microsoft’s Data
Warehouse Fast Track certification program is a great way to determine which combination of hardware and
storage deliver the most value for your business.

Audience
This reference architecture is targeted toward Enterprise Architects, CIOs, CTOs, technical managers, and
data professionals such as BI Architects, BI Administrators and Database Administrators.

Pure Storage Introduction

Who knew that moving to all-flash storage could help reduce the cost of IT? FlashArray//m makes server and
workload investments more productive, while also lowering storage spend. With FlashArray//m,
organizations can dramatically reduce the complexity of storage
to make IT more agile and efficient, accelerating your journey to
the cloud.

FlashArray//m’s performance can also make your business smarter

by unleashing the power of real-time analytics, driving customer
loyalty, and creating new, innovative customer experiences that
simply weren’t possible with disk. All by Transforming Your
Storage with FlashArray//m.

FlashArray//m enables you to transform your data center, cloud, or

entire business with an affordable all-flash array capable of consolidating and accelerating all your key
applications.

Mini Size—Reduce power, space and complexity by 90%

• 3U base chassis with 15-120+ TBs usable

© Pure Storage 2015 | 5

 • ~1kW of power

• 6 cables

Mighty Performance—Transform your datacenter, cloud, or entire business

• Up to 300,000 32K IOPS

• Up to 9 GB/s bandwidth

• <1ms average latency

Modular Scale—Scale FlashArray//m inside and outside of the chassis for generations

• Expandable to ~½ PB usable via expansion shelves

• Upgrade controllers and drives to expand performance and/or capacity

Meaningful Simplicity—Appliance-like deployment with worry-free operations

• Plug-and-go deployment that takes minutes, not days

• Non-disruptive upgrades and hot-swap everything

• Less parts = more reliability

The FlashArray//m expands upon the FlashArray’s modular, stateless architecture, designed to enable
expandability and upgradability for generations. The FlashArray//m leverages a chassis-based design with
customizable modules, enabling both capacity and performance to be independently improved over time
with advances in compute and flash, to meet your business’ needs today and tomorrow.

The Pure Storage FlashArray is ideal for:

Accelerating Databases and Applications Speed transactions by 10x with consistent low latency, enable online
data analytics across wide datasets, and mix production, analytics, dev/test, and backup workloads without
fear.

Virtualizing and Consolidating Workloads Easily accommodate the most IO-hungry Tier 1 workloads, increase
consolidation rates (thereby reducing servers), simplify VI administration, and accelerate common
administrative tasks.

Delivering the Ultimate Virtual Desktop Experience Support demanding users with better performance than
physical desktops, scale without disruption from pilot to >1000’s of users, and experience all-flash
performance for under $100/desktop.

Protecting and Recovering Vital Data Assets Provide an always-on protection for business-critical data,
maintain performance even under failure conditions, and recover instantly with FlashRecover.

Pure Storage FlashArray sets the benchmark for all-flash enterprise storage arrays. It delivers:

© Pure Storage 2015 | 6

Consistent Performance FlashArray delivers consistent <1ms average latency. Performance is optimized for
the real-world applications workloads that are dominated by I/O sizes of 32K or larger vs. 4K/8K hero
performance benchmarks. Full performance is maintained even under failures/updates.

Less Cost than Disk Inline de-duplication and compression deliver 5 – 10x space savings across a broad set of
I/O workloads including Databases, Virtual Machines and Virtual Desktop Infrastructure.

Mission-Critical Resiliency FlashArray delivers >99.999% proven availability, as measured across the Pure
Storage installed base and does so with non-disruptive everything without performance impact.

Disaster Recovery Built-In FlashArray offers native, fully-integrated, data reduction-optimized backup and
disaster recovery at no additional cost. Setup disaster recovery with policy-based automation within minutes.
And, recover instantly from local, space-efficient snapshots or remote replicas.

Simplicity Built-In FlashArray offers game-changing management simplicity that makes storage installation,
configuration, provisioning and migration a snap. No more managing performance, RAID, tiers or caching.
Achieve optimal application performance without any tuning at any layer. Manage the FlashArray the way
you like it: Web-based GUI, CLI, VMware vCenter, Rest API, or OpenStack.

Table 1. Pure Storage FlashArray//m Series.

FlashArray//m Specifications
//m20 //m50 //m70
Capacity • Up to 120+ TBs effective capacity* • Up to 250+ TBs effective capacity* • Up to 400+ TBs effective capacity*
• 5 – 40TBs raw capacity • 30 – 88TBs raw capacity (w/shelves) • 44 – 136TBs raw capacity
(base chassis) (w/shelves)

Performance • Up to 150,000 32K IOPS** • Up to 220,000 32K IOPS** • Up to 300,000 32K IOPS**
• <1ms average latency • <1ms average latency • <1ms average latency
• Up to 5 GB/s bandwidth • Up to 7 GB/s bandwidth • Up to 9 GB/s bandwidth

Connectivity • 8 Gb/s Fibre Channel • 16 Gb/s Fibre Channel • 16 Gb/s Fibre Channel
• 10 Gb/s Ethernet iSCSI • 10 Gb/s Ethernet iSCSI • 10 Gb/s Ethernet iSCSI
• Management and Replication ports • Management and Replication ports • Management and Replication ports

© Pure Storage 2015 | 7

Physical • 3U • 3U – 7U • 5U – 11U
• 742 Watts (nominal draw) • 1007 - 1447 Watts (nominal draw) • 1439 – 2099 Watts (nominal draw)
• 110 lbs. (49.9 kg) fully loaded • 110 lbs. (49.9 kg) fully loaded + 44 • 110 lbs. (49.9 kg) fully loaded + 44
• 5.12” x 18.94” x 29.72” lbs. per expansion shelf lbs. per expansion shelf
FlashArray//m chassis • 5.12” x 18.94” x 29.72” • 5.12” x 18.94” x 29.72”
FlashArray//m chassis FlashArray//m chassis

* Effective capacity assumes HA, RAID, and metadata overhead, GB-to-GiB conversion, and includes the benefit of data
reduction with always-on inline deduplication, compression, and pattern removal. Average data reduction is calculated at 5-to-1,
below the global average of the FlashArray user base.

** Why does Pure Storage quote 32K, not 4K IOPS? The industry commonly markets 4K IOPS, but real-world environments are dominated by
IO sizes of 32K or larger. FlashArray//m adapts automatically to 512B-32KB IO for superior performance, scalability, and data reduction.

Purity Operating Environment

Purity implements advanced data reduction, storage management and flash management features, and all
features of Purity are included in the base cost of the FlashArray//m.

Storage Software Built for Flash—The FlashCare technology virtualizes the entire pool of flash within the
FlashArray, and allows Purity to both extend the life and ensure the maximum performance of consumer-
grade MLC flash.

Granular and Adaptive—Purity Core is based upon a 512-byte variable block size metadata layer. This fine-
grain metadata enables all of Purity’s data and flash management services to operate at the highest
efficiency.

Best Data Reduction Available—FlashReduce implements five forms of inline and post-process data
reduction to offer the most complete data reduction in the industry. Data reduction operates at a 512-byte
aligned variable block size, to enable effective reduction across a wide range of mixed workloads without
tuning.

Highly Available and Resilient—FlashProtect implements high availability, dual-parity RAID-3D, non-
disruptive upgrades, and encryption, all of which are designed to deliver full performance to the FlashArray
during any failure or maintenance event.

Backup and Disaster Recovery Built In—FlashRecover combines space-saving snapshots, replication, and
protection policies into an end-to-end data protection and recovery solution that protects data against loss
locally and globally. All FlashProtect services are fully-integrated in the FlashArray and leverage the native
data reduction capabilities.

Pure1

Pure1 Manage—By combining local web-based management with cloud-based monitoring, Pure1 Manage
allows you to manage your FlashArray wherever you are – with just a web browser.

Pure1 Connect—A rich set of APIs, plugin-is, application connectors, and automation toolkits enable you to
connect FlashArray//m to all your data center and cloud monitoring, management, and orchestration tools.

Pure1 Support—FlashArray//m is constantly cloud- connected, enabling Pure Storage to deliver the most
proactive support experience possible. Highly trained staff combined with big data analytics help resolve
problems before they start.

© Pure Storage 2015 | 8

Pure1 Collaborate—Extend your development and support experience online, leveraging the Pure1
Collaborate community to get peer-based support, and to share tips, tricks, and scripts.

Experience Evergreen Storage

Tired of the 3-5 year array replacement merry-go-round? The move to FlashArray//m can be your
last data migration. Purchase and deploy storage once and once only – then expand capacity and
performance incrementally in conjunction with your business needs and without downtime. Pure
Storage’s vision for Evergreen Storage is delivered by a combination of the FlashArray’s stateless,
modular architecture and the ForeverFlash business model, enabling you to extend the lifecycle of
storage from 3-5 years to a decade or more.

About the SQL Server Data Warehouse Fast Track Program

Through the Data Warehouse Fast Track Reference Architecture program, Microsoft aims to provide end
users with validated architectures that are required to achieve and maintain a certain level of performance
with minimal configuration.

A Data Warehouse Fast Track Reference Architecture helps businesses:

• Reduce Total Cost of Ownership (TCO), since the architecture has been pre-tested and adheres to
Microsoft’s guidelines for predictable, consistent performance.

• Reduce time to go-live for data warehousing projects.

• Reduce the complexity of designing and maintaining a data warehousing solution.

About the Solution

Pure Storage’s proposed solution for a 20TB data warehouse is based on a FlashArray//m20 array, which is a
very cost-effective entry-level array that can scale non-disruptively to almost half a petabyte of usable storage
and over 300,000 32K-sized IOPS at less than 1 millisecond latency, and 10 Gigabytes per second of
throughput.

The array is connected directly to a Cisco C240 M4 server via eight (8) 8 gigabit-per-second fibre channel
ports on four (4) dual-port host bus adapters*. This maximizes throughput available, which is essential to
data warehousing workloads.

The server used in the solution is an off-the-shelf one, with 256GB of RAM. Cost-effective CPUs were used in
this solution (Intel Xeon E5-2670 v3). The server has dual CPU sockets, and each CPU has 24 hyper threaded
cores, for a total of 48 logical processors in the system.

The entire solution fits in 5 rack units: 3 units for the FlashArray//m20, 2 rack units for the Cisco C240 M4
server.

© Pure Storage 2015 | 9

 Although the HBAs used support 16Gbps connectivity, we used 8Gbps SFP+ modules for this
solution. The FlashArray//m20 supports only 8Gbps FC connectivity, while the //m50 and
//m70 support 16Gbps FC connectivity. Scalability, plenty of room to grow – and you can
grow while keeping your data warehouse completely online.

Windows Server Configuration Settings

The changes made to Windows Server’s default configuration for this solution are straightforward. Following
Pure Storage’s Windows Server best practices, the MPIO feature was installed, and we ensured Pure volumes
had the appropriate multi-pathing load balancing policy applied. We also set the Power Plan to High
Performance, and set the MPIO path timeout and failover options per our MPIO Best Practices.

Storage Layout
The storage layout for this solution is extremely simple. As our reference server had no direct attached
storage, we configured it to boot from SAN on a FlashArray volume. We also created four volumes for it, in
addition to the boot volume:

1. Anchor (10GB) – a small volume that serves as the root and contains mount points for all other
volumes
2. tempdb (1TB) – placed all tempdb database files and transaction log file on this volume
3. Data and log files (8TB) – placed all user database files and transaction log files on this volume
4. Backup (2TB) – created a volume for backups (not required by the DWFT certification, however we do
suggest backups be split onto their own volume per our best practices)

Data Reduction on Compressed Databases

Thanks to FlashArray’s unmatched data reduction capabilities, the RowStore Data Warehouse Fast Track
reference database storage footprint is reduced by a factor of 2:1. Note that this level of data reduction is
achieved on top of the reduction achieved from PAGE-compressed tables. Windows Server sees an 8TB
provisioned volume, with database and transaction log files measuring at 1.63TB. And, because of the Purity
Operating Environment’s strong security features, the data is automatically encrypted at rest without
compromising performance.

© Pure Storage 2015 | 10

The actual storage footprint for this database is roughly 608GB.

Similarly, the ColumnStore-based version of the database reduces nicely to 1.7:1, occupying 913GB of space in
the array.

© Pure Storage 2015 | 11

SQL Server Configuration Settings
In this section we will provide details around the SQL Server database engine configuration settings used for
this solution. We will break them down into three areas:

1. General instance settings and trace flags

2. Resource Governor configuration

3. Database-specific settings

General Settings and Trace Flags

Here are the configuration settings and trace flags used by the SQL Server instance:

Setting or Trace Flag Value

Number of tempdb files 12x 2GB files

Maximum Degree of 24, however the Fast Track utility can push MAXDOP as high as
Parallelism (MAXDOP) the number of logical cores in the system

Max Server Memory (MB) 120832

-E startup option (Allocate Enabled

multiple extents per file in
the filegroup)

Trace Flag 834 (Use large Enabled as startup parameter

pages for memory used
for buffer pool)

Trace Flag 1117 (Grow data Enabled as startup parameter

files uniformly upon
autogrowth event)

Trace Flag 1118 (Disabled Enabled as startup parameter

shared extents)

Resource Governor Configuration

The Data Warehouse Fast Track guidelines call for a 12% maximum memory grant for all queries submitted to
the instance while certification tests take place. The intention is to increase concurrency of the system by
preventing any one query from stalling another query’s memory grant request. The setting takes effect for
all queries submitted to the system by modifying the “default” workload group. No other modifications were
made to Resource Governor’s configuration.

Database-Specific Settings

© Pure Storage 2015 | 12

The Data Warehouse Fast Track certification performs two sets of tests: a RowStore test, where all indexes
and heaps for every single table in the reference database are compressed with PAGE compression, and a
ColumnStore test, where every single table in the reference database has a Clustered ColumnStore index.

In this solution none of these compression settings were tuned or modified in any way, shape, or form. Pure
Storage FlashArrays data-reduction arrays can deliver excellent performance on both compressed and
uncompressed data.

Solution Hardware & Software Overview

Let us look at the hardware and software components of this solution. This solution requires a minimal set of
hardware and software:

Hardware
Cisco C240 M4 server with 2x Intel Xeon E5-2670 v3 CPUs, 256GB of RAM

Pure Storage FlashArray//m20, 20TB raw capacity (2x 10TB Capacity Packs), 4x Dual-port 8Gbps FC HBAs

Software
Purity Operating Environment 4.5.8

Microsoft Windows Server 2012 R2 Standard Edition

Microsoft SQL Server 2014 Enterprise Edition with Service Pack 1

Bill Of Materials For Cisco C240 M4 Server

This is the bill of materials that details all the parts used in the Cisco C240 M4 server for this solution:

Part Name Quantity Description

UCS-SPL-C240M4-A2 1 UCS SP Select C240M4SX Advanced2 w/2xE52670v3, ,8x32GB

UCS-CPU-E52670D 2 2.30 GHz E5-2670 v3/120W 12C/30MB Cache/DDR4 2133MHz

UCS-ML-1X324RU-A 8 32GB DDR4—2133—MHz LRDIMM/PC4—17000/quad

rank/x4/1.2v

UCSC-PCI-1-C240M4 1 Right PCIe Riser Board (Riser 1) for C240 M4

UCSC-PCI-2-C240M4 1 Left PCIe Riser Board (Riser 2) for C240 M4

UCSC-PSU2-1400W 1 1400W AC Power Supply for 2U & 4U C Series Servers

UCSC-RAILB-M4 1 Ball Bearing Rail Kit for C220 M4 and C240 M4 rack sewers

N20-BBLKD 24 UCS 2.5 inch HDD blanking panel

© Pure Storage 2015 | 13

 N20-BKVM 1 KVM local IO cable for UCS servers console port

UCSC-HS-C240M4 1 Heat sink for UCS C240 M4 rack servers

UCSC-SCCBL240 1 Supercap cable 250mm

CAB-9K12A-NA 2 Power Cord, 125VAC 13A NEMA 5-15 Plug, North America

QLE2672-CK 4 QLogic QLE2672-CK – dual port 16Gbps FC host bus adapter

Table 1. Bill of Materials for Cisco C240 M4 Server

Certification
The figure below shows the Microsoft SQL Server Data Warehouse Fast Track certification awarded to our
FlashArray//m20-based solution described in this document. Of particular interest, the columns “Measured
Scan Rate Physical” and “Measured I/O Throughput”.

© Pure Storage 2015 | 14

Figure 1. Data Warehouse Fast Track Certification

© Pure Storage 2015 | 15

Summary
The test results highlighted in this document clearly demonstrate the high performance, resiliency and
scalabilty of Pure Storage’s entry level FlashArray//m20 for Microsoft SQL Server Data Warehouse
environments. We also demonstrated how customers can achieve these benefits with minimal tuning and
configuration. These benefits will be even more pronounced on the higher level FlashArray //m50 and //m70.

References
In this section you will find some relevant links for further information on this solution:

• SQL Server on Pure Storage: Best Practices (Pure Storage Community, login required)
http://community.purestorage.com/t5/Interoperability-Best-Practice/Microsoft-SQL-Server-Best-
Practices-Community-Page/ta-p/6097

• Fast Track Data Warehouse Reference Guide for SQL Server 2012 – Microsoft
https://msdn.microsoft.com/en-us/library/hh918452.aspx

• Cisco UCS C240 M4 Server Installation and Service Guide

http://www.cisco.com/c/en/us/td/docs/unified_computing/ucs/c/hw/C240M4/install/C240M4.html

• Windows Server on Pure Storage: Best Practices (Pure Storage Community, login required)
http://community.purestorage.com/t5/Interoperability-Best-Practice/Windows-Server-Best-
Practices/ta-p/166

© Pure Storage 2015 | 16

About the Author
Argenis Fernandez is a SQL Server Solutions Architect for Pure Storage
based in the Seattle, Washington area. He has worked with SQL Server for
around 17 years and enjoys large SQL Server environments, high-end OLTP
databases, performance troubleshooting, high availability, disaster
recovery, best practices, and PowerShell scripting.

Prior to Pure Storage, Argenis led the Database Engineering team at

SurveyMonkey, where he successfully replaced the database storage
infrastructure with all-flash technologies from Pure Storage. Argenis has
also worked as a Senior Consultant on SQL Server Core Database Engine
for Microsoft.

Argenis has been a Microsoft Most Valuable Professional (MVP) for Data
Platform (previously known as SQL Server MVP) since 2014 – an award
given to highly influential technologists in the data management and
analytics space.

Argenis is a SQL community enthusiast and speaks frequently at major SQL Server conferences, including the
PASS Summit, IT/Dev Connections, SQLBits, and TechEd. He has also spoken at many SQLSaturday events
around the globe. He is also a Microsoft Certified Master (MCM), the highest technical certification Microsoft
ever offered, with less than 200 certified individuals worldwide. In 2013 he founded the Security Virtual
Chapter for the Professional Association for SQL Server (PASS) (http://security.sqlpass.org) .

Blog: http://blog.purestorage.com/blog/author/argenis

Twitter: @DBArgenis

© Pure Storage 2015 | 17

 Pure Storage, Inc.
Twitter: @purestorage
www.purestorage.com

650 Castro Street, Suite #260

Mountain View, CA 94041

T: 650-290-6088
F: 650-625-9667

Sales: [email protected]
Support: [email protected]
Media: [email protected]
General: [email protected]

© Pure Storage 2015 | 18

© 2015 Pure Storage, Inc. All rights reserved. Pure Storage, Pure1, and the P Logo are trademarks of Pure Storage, Inc. All other trademarks are the property of their respective owners.