Citrix XenDesktop Reference Architecture for 750 users

High Performance 750-Seat Virtual
Desktop Infrastructure (VDI)
with Citrix XenDesktop 7
Reference Architecture
February, 2014

2 High-Performance 750-Seat VDI Reference Architecture powered by X-IO and Citrix
Table of Contents
Executive Summary ...................................................................................................... 3
Infrastructural Components ......................................................................................... 4
Citrix XenDesktop .....................................................................................................................................4
Microsoft Windows Server 2012 R2 Hyper-V ...........................................................................................4
The X-IO Intelligent Storage Element .......................................................................................................5
Login VSI...................................................................................................................................................6
Reference Architecture Configuration ........................................................................ 7
Physical Hardware Configuration..............................................................................................................7
Virtual Machine Configuration...................................................................................................................8
Citrix XenDesktop Storage Requirements ................................................................................................9
Windows Server 2012 R2 Configuration and Best Practices....................................................................9
Network Configuration.............................................................................................................................10
Intelligent Storage Element Test Configuration ....................................................... 11
Fibre Channel Switch Configuration .......................................................................................................11
LUN Configuration...................................................................................................................................12
Login VSI Configuration ..........................................................................................................................13
Test Results................................................................................................................. 14
ISE 230 Test Results ..............................................................................................................................14
Login VSI Test Results............................................................................................................................17
Total Cost of Ownership (TCO) Analysis .................................................................. 19
TCO versus Cost of Acquisition ..............................................................................................................19
Sizing Storage for Actual Performance Requirements ...........................................................................19
Storage Density and Rack Space Consumption.....................................................................................20
Storage Energy Consumption.................................................................................................................20
Storage Warranty Costs..........................................................................................................................22
Summary...................................................................................................................... 22
Acknowledgements..................................................................................................... 23
Thin Client Computing.............................................................................................................................23
Citrix ........................................................................................................................................................23
Contact Information.................................................................................................................................23
Appendix A: ISE 230 Tested Configuration Details.................................................. 24

Executive Summary
This document describes a fully-tested, benchmarked and proofed reference architecture for large-scale
(750+ user) virtual desktop infrastructure (VDI) implementations.
The target audience for this document includes storage and virtualization administrators, consulting data
center architects, field engineers, cloud architects, and desktop specialists who want to implement Citrix
XenDesktop based virtual desktops. A working knowledge of Hyper-V, XenDesktop, server, storage,
network and data center design is assumed but is not a prerequisite to read this document.
VDI (Virtual Desktop Infrastructure) enables IT to deliver a secure, mobile, manageable and less costly
end user computing platform, at lower cost. That is the essential promise of VDI.
However a large number of VDI projects fail to deliver on that promise, in most cases due to storage
performance bottlenecks. As a result:
 The VDI project cannot be completed, successfully at cost.
 The performance of the VDI environment is significantly worse than the performance of dedicated-
hardware desktops.
 The cost of storage sufficient to equalize the VDI performance deficit makes the cost of the VDI project
greater than the cost of dedicated-hardware desktops.
 The storage performance bottlenecks that cripple so many VDI projects are the result, in some cases,
of a failure to characterize the per-VM performance requirements for a VDI environment, but are more
often the result of a damaging, single-minded focus on lower cost storage capacity ($/GB).
 In our view, it is absolutely critical that VDI projects balance both storage performance and storage
capacity in order to be successful.
When per-VM storage performance requirements are taken into consideration along with per-VM storage
capacity requirements, during the planning phase of VDI projects:
- Traditional enterprise storage array vendors are forced to over-provision, or short-stroke, storage
capacity to deliver enough storage performance for the required number of VMs, producing a very
expensive solution.
- Flash and solid state disk (SSD) array vendors wildly over-provision storage performance to
deliver enough storage capacity to each VM.
This reference architecture and its underlying results demonstrate, without a doubt, the suitability of an
all-hard disk drive (HDD) ISE configuration without the need for higher cost all flash or hybrid arrays.

Infrastructural Components
The tested VDI reference architecture described in this document is based on the following:
 Citrix XenDesktop 7
 Microsoft Windows Server 2012 R2 Hyper-V
 X-IO Intelligent Storage Elements (ISEs) and ISE Manager software
 Login VSI
Citrix XenDesktop
Citrix XenDesktop delivers Windows apps and desktops as secure mobile services. With XenDesktop, IT
can mobilize the business, while reducing costs by centralizing control and security for intellectual
property. Incorporating the full power of XenApp, XenDesktop can deliver full desktops or just the apps to
any device. HDX technologies enable XenDesktop to deliver a native touch-enabled look-and-feel that is
optimized for the type of device, as well as the network.
Citrix XenDesktop provides the best method available today to provision, host, manage and deliver virtual
desktops and applications. Citrix has the benefit of 25 years of advanced expertise in optimizing remote
protocols and virtualizing user workloads.
User experience is greatly enhanced with Citrix XenDesktop by providing high performance, 24x7 access
to all corporate apps and data, from anywhere, from any type of device (PC, Mac, Linux, Tablets, IOS,
Android, etc.)
For the IT department, Citrix centralizes management in dramatically more efficient ways. Rather than
maintaining the legacy approach of installing and managing operating systems, applications, patches and
customizations on every single device, Citrix allows a “One to Many” solution. A single desktop image can
service 100’s, 1000’s, or, 10’s of thousands of users providing a single point of update for the entire IT
environment. At the same time, security and performance are greatly improved by keeping the
applications close to the data in the secure boundary of the data center.
Microsoft Windows Server 2012 R2 Hyper-V
Hyper-V is an important part of Windows Server and provides a highly developed virtualization platform
that lets cloud providers provision for customers, and enterprises build their own private cloud of hosted
desktops. With Windows Server 2012 R2, customers get a compelling solution for core virtualization
scenarios; production server consolidation, dynamic data center, business continuity, test and
development and most importantly, as it relates to this reference architecture, Virtual Desktop
Infrastructure (VDI). Hyper-V provides customers with better flexibility with features like live migration and
cluster shared volumes for storage flexibility. In Windows Server 2012 R2, Hyper-V also delivers greater
scalability and improved performance with support for dynamic memory and enhanced networking
support versus other competing hypervisors or previous versions of Hyper-V.
Virtualization technologies help customers lower costs and deliver greater agility and economies of scale.
Either as a stand-alone product or an integrated part of Windows Server, Hyper-V is the leading
virtualization platform for today and the transformational opportunity with cloud computing, and hosted
desktops.

The X-IO Intelligent Storage Element
X-IO’s Intelligent Storage Elements (ISEs) – both the 700 series hybrid storage arrays and the 200 series
all-hard disk drive storage arrays – are frequently deployed for large-scale VDI projects because:
 ISEs offer predictably high levels of VDI performance. Performance is critical to the success of VDI
implementations. It impacts everything: user experience, application performance and the overall cost of
the environment.
 ISEs scale linearly. The ISE can support thousands of virtual desktops with a “scale-out” architecture
design that provides modular, linear scaling as needed for VDI deployments.
 ISEs offer industry-leading VDI price/performance. The ISE’s balance of performance, capacity and price
significantly improves the economics of large-scale virtual desktop environments, when compared to
conventional enterprise storage arrays.
 ISEs offer best-in-class reliability. ISEs are highly available, supporting today’s high-availability enterprise
applications reliability demands with 99.999% uptime. Redundancy at every level of the ISE ensures that
if a hardware failure occurs, all application workloads will continue unaffected. X-IO Technologies is so
confident in the ISE’s technical and operational reliability that it warranties the ISE for five years at no cost
to its customers.
 ISEs are easy to deploy. Traditional storage systems are complex and difficult to troubleshoot, when
adjusting performance in VDI deployments. By contrast, the ISE Manager software, an intuitive storage
management environment that provides simple, user-friendly storage and integrated server
administration, allows virtualization administrators to provision and troubleshoot the entire VDI “stack”
through a single, easy-to-use graphical interface.
Figure 1 – ISE Manager Storage View

Login VSI
Login VSI is and industry standard load testing tool for Virtual Desktop Infrastructures and Server Based
Computing environments.
With Login VSI you can test the performance and scalability of Citrix XenDesktop and Citrix XenApp, by
simulating unique user workloads.
With Login VSI you can:
 Benchmark different desktop, server and storage solutions and make the right decisions.
 Know how much hardware you need by right sizing your XenDesktop infrastructure.
 Gain insight in the maximum capacity by load testing your XenDesktop environment.
 Test the impact of changes on the performance before they are implemented.

Reference Architecture Configuration
The system-under-test (SUT) was configured as described below.
Physical Hardware Configuration
 One Hewlett-Packard C7000 blade enclosure with 16 BL460 G7 servers, Intel Xeon X5650 (2.67
GHz, 6 cores), 192 GB memory, Windows Server 2012
o Hyper-Threading was enabled in the BIOS of all 16 blade servers
o Max Performance was enabled in the BIOS of all 16 blade servers
o Two BL460 Servers were configured for Infrastructure Services
o Three BL460 Servers were configured for Login VSI Client Workload Launchers
o Eleven BL460 Servers were configured for hosted desktop virtualization
 One QLogic 5802v Fibre Channel switch
 Two X-IO ISE 230 storage systems
o Eleven write cache file LUNs; one LUN per XenDesktop blade server
o Eleven user data LUNs, one LUN per XenDesktop blade server
Figure 2 shows the detailed topology of the reference architecture including the HP Blade enclosure
with server role allocations, SAN topology and X-IO ISE storage arrays.
Figure 2 - Detailed Reference Architecture Configuration

Virtual Machine Configuration
Infrastructure Services:
 Windows Server 2012 Hyper-V
 Windows Server 2012 Virtual Machines
o Active Directory, DHCP and DNS – 4 GB memory, 1 vCPU
o Citrix Provisioning Services (PVS) 7.0 running SQL Server – 4 GB memory, 2 vCPU
o Citrix Desktop Delivery Controller (DDC) running SQL Server – 8 GB memory, 2 vCPU
o Citrix License server VM
Login VSI client launchers:
 Windows Server 2012 Hyper-V
 Windows 7 32-bit Virtual Machines – 1.5 GB memory, 1 vCPU
XenDesktop Server:
 Windows Server 2012 R2 Hyper-V with MPIO feature added
 Windows 7 64-bit Virtual Machines – 1.5 GB memory, 1 vCPU
Virtual Machine Configuration Diagram
Figure 3 shows a topology of the tested environment for this reference architecture.
The Citrix XenDesktop 7 infrastructure was configured on dedicated hosts. The infrastructure VMs were
hosted on local server storage.
Figure 3 - Test Environment overview of Citrix XenDesktop
deployment with Infrastructure components, Windows
Server 2012 R2 Hyper-V hosts and ISE volumes

Citrix XenDesktop Storage Requirements
XenDesktop FlexCast Model has two different data types (write cache files and user data) with completely
different storage performance criteria.
The following table details the tested ISE per-user storage configuration:
Data Type Per user capacity
Write cache files 8GB/user x 750 users (6TB total)
User data 41.7GB/user x 750 users (31.3TB total)
Total 49.7GB/user x 750 users (37.3TB)
X-IO Tested Storage Configuration
Windows Server 2012 R2 Configuration and Best Practices
The Windows Server 2012 R2 Hyper-V role and MPIO feature were added using the following PowerShell
command:
# Install-WindowsFeature Hyper-V, Hyper-V-PowerShell, Hyper-V-Tools, Multipath-IO
The Windows Server power plan was set to “High performance” using the following PowerShell
command:
# POWERCFG.EXE /S SCHEME_MIN
The ISEs utilized the Windows Server 2012 R2 Multi-Path IO Microsoft Device Specific Module (MS
DSM).
Native DSM:
 Does not automatically claim enterprise class storage device LUNs, by default
 Uses a round-robin load balancing policy, by default
The tested MPIO configuration added “Xiotech ISE2400” to native DSM managed devices. MPIO Policy
was set to Least Queue Depth (LQD).
The following mpclaim commands were used, via PowerShell, to set all MPIO load balance policy for
“XIOTECH ISE2400” for all LUNs.
# List enterprise storage device for vendor string
mpclaim –e
# List all MPIO disks controlled by MSDSM (native DSM)
mpclaim –s –d
# Add multipath support for device
# Suppress reboot
mpclaim –n –I –d “XIOTECH ISE2400”
# Set load balance policy to LQD for all disks controlled by MSDSM
mpclaim –l –m 4

# Restart server with reconfigured MPIO settings
shutdown –r –t 00
Network Configuration
Network 4 was configured to allow for Fibre connectivity from the HP blade servers. These were
connected to an 8Gb QLogic 5802v.
 Four networks will be created:
o Network 1 – Internal to HP Virtual Connect for PXE boot of VMs, allocated 5Gb bandwidth
o Network 2 – Lab storage and management, 1Gb bandwidth
o Network 3 – Production network for workload clients and XenDesktop VMs, 5Gb bandwidth
o Network 4 – Fibre connection to ISE storage arrays, 8Gb bandwidth
Note: There will be no HA or redundancy across the NICs
Figure 4 illustrates the HP Blade Enclosure internal network design and external storage network.
Figure 4 - Network Configuration

Intelligent Storage Element Test Configuration
Two ISE 230 systems providing eighty (80) 900 GB, 10K RPM SAS hard disk drives rated at 57.6 TB
usable capacity and 27,000 IOPS were used for this tested reference configuration. The ISEs provided
that performance and capacity in 6U of rack space, while consuming less than 1.5 kilowatts (kW) of
energy. The ISEs were not tuned in any way before or during the tests described in the next section.
There are four models in the ISE 200 series, each delivering real-world performance of 13,500 IOPS and
between 9.6 and 38.4TB of usable capacity in each 3U array.
Number of
ISE
Rated IOPS
ISE 210
Useable
Capacity
ISE 220
Useable
Capacity
ISE 230
Useable
Capacity
ISE 240
Useable
Capacity
1 13,500 IOPS 9.6 TB 19.2 TB 28.8 TB 38.4 TB
2 27,000 IOPS 19.2 TB 38.4 TB 57.6 TB 76.8 TB
For this reference architecture, the storage configuration must deliver a mix of aggregate performance at
a minimum of 25,000 IOPS (30 IOPS per defined user VM) and useable capacity (8 GBs/user for write
cache files and 41.7 GB/user for user data): easily within the rated performance and rated capacity of a
pair of ISE 230s.
Fibre Channel Switch Configuration
A single QLogic 5802v 8Gb Fibre Channel (FC) switch with 8 licensed ports was installed per established
QLogic guidelines and user documentation. The Fibre Channel zoning was configured to be open.
For smaller number of VDI users, it is possible to remove the Fibre Channel switch from the configuration
completely and connect up to four servers directly to a single ISE via the server’s Fibre Channel host bus
adapter.
For high availability configurations, a second QLogic 5802v 8Gb FC switch is required for VDI
configurations requiring multiple ISE arrays and/or multiple servers.

LUN Configuration
In the tables below, the ISEs are identified by unit letter (A or B), by use (L for write cache file and D for
user data) and blade number (blade 01 – 11). For example, “ISE-A-L01” denotes a LUN on ISE A, used
for write cache, and mapped to Blade 1.
ISE-A
 LUN Name  Data Use  RAID Type  LUN Size  Server
 ISE-A-L01  Write cache  ISE RAID-10  551 GB  Blade 01
 ISE-A-D01  User data  ISE RAID-5  3001 GB  Blade 01
ISE-B
LUN Name Data Use RAID Type LUN Size Server
ISE-B-L07 Write cache ISE RAID-10 557 GB Blade 07
ISE-B-D06 User data ISE RAID-5 3006 GB Blade 06

Login VSI Configuration
Login VSI is designed to perform benchmarks for VDI workloads through system saturation. Login VSI
loads the system with simulated user workloads using well known desktop applications such as Microsoft
Office, Internet Explorer and Adobe PDF reader.
Login VSI is the de-facto industry standard benchmarking tool to test the performance and scalability of
centralized Windows desktop environments. Login VSI is 100% vendor independent and is used to test
virtual desktop environments like Citrix XenDesktop.
Medium Workload
This workload is based on the standard Login VSI medium knowledge worker workload, employing a mix
of Microsoft Office applications, Internet Explorer, PDF conversion software and 7-zip. Once a test
session has been started, the workload will repeat (loop) every 12 minutes. This loop is divided in
segments, and each consecutive Login VSI user logon will start in a different segment. This ensures that
all elements in the workload are equally used throughout the test. The medium workload opens up to 5
applications simultaneously, with the keyboard type rate set at 160ms for each character. During each
loop the response time is measured every 2 minutes with only 2 minutes of idle time to simulate real‐
world users.

Test Results
This reference architecture was tested in November 2013, at the Citrix Labs in Santa Clara, California, to
verify its ability to support at least 750 simultaneous VDI sessions/instances.
Two sets of tests were performed in each pass of the testing process:
Test 1 – Boot Storm Test: 750 user VMs must boot and register with a VSI broker within 10 minutes, with
boot start occurring at the rate of 75 VMs per minute
Test 2 – Login VSI “Medium” Workload Test: 750 user VMs must run the Login VSI “medium” knowledge
worker workload, with average response time latency not exceeding 3811 milliseconds (Baseline 1211 +
Tolerance 2600 = Threshold 3811), also known as VSImax.
Three test runs were performed on the configuration described in the previous section. Each test run did
not hit VSImax and was marked Pass; none of the test runs resulted in saturation or exceeded the
response time latency threshold (in this test, 3811 milliseconds), and no configuration changes were
made to the ISE storage between test passes. Server changes were made between test passes to
increase the number of active VDI sessions achievable with the configuration.
Detailed storage performance information was recorded, during each pass, with the ISE Performance
Adapter for Windows Performance Monitor. This tool, included with ISE Manager, allows for extensive
performance data capture and analysis using the Windows Performance Monitor Framework.
ISE 230 Test Results
Boot Storm Test
Figure 5 is the main performance graph of both ISEs during the “Boot Storm” phase of the testing.
Figure 5 - Boot Storm Read and Write IOPS
During this phase, the ISE storage systems can be seen regularly servicing IO spikes over 20,000 IOPS,
with peaks of over 30,000 IOPS. Write IOs comprise the vast majority of the workload, which the ISE

arrays handled easily. At peak times, the pair of ISEs delivered approximately 375 IOPS per hard disk
drive.
In figure 6, most of the Read Latencies during the boot storm test were below 3ms. Latency below 10 ms
is considered good, in a VDI environment, and the ISEs lower latencies indicate that the system is
performance-optimized for the workload.
Figure 6 - Boot Storm Read Latency
In figure 7, the Write Latencies during the Boot Storm test were below 1ms. Again, the ISEs were easily
able to handle the workload.
Figure 7 - Boot Storm Write Latency

Login VSI Medium Workload Test
In Figure 8, the vast majority of the workload is shown to be heavily write-oriented (or “write-intensive”) in
nature. Writes peaked, during this test, at over 7,000 per second, and were routinely at or above the
6000 IOPS level.
Figure 8 - User Load Read and Write Total IOPS
In Figure 9, 95% of the read latency values were below 3 milliseconds, with a single observed high value
of 10ms, indicating clearly that the ISEs were having no difficulty servicing the read-oriented portion of the
workload.
Figure 9 - User Load Average Read Latency

Figure 10 shows write latency on the ISE arrays under test. Any latency values below 5-10 ms are
considered good. The write latency results below are excellent and show the ISE arrays are able to
service these workloads with no signs of reaching a performance limit.
Figure 10 - User Load Average Write Latency
The data recorded from the ISE arrays above demonstrate a high performance, write-intensive workload.
The ISE arrays easily handle this intensive workload as evident by the sub-millisecond latency response
times. This is very important to consider as many high performance storage products on the market are
optimized for random, read-intensive operations, through the use of a flash-enabled read-cache, for
example.
Login VSI Test Results
The Login VSI VSImax chart below documents the actual number of maximum virtual desktop sessions
supported by the configuration.

Figure 11 - VSImax
Figure 11 shows Login VSI v3.7 VSImax is not reached, resulting in a successful test. With Read
Latencies mostly at 3ms and Write Latencies at 1ms, the ISE arrays are carrying the workload of the 750
XenDesktop VMs without issue.

Total Cost of Ownership (TCO) Analysis
The test results for this reference architecture demonstrate clearly that the configuration is a high-
performance solution for large-scale VDI workload requirements. The documented reference architecture
is also a superior economic alternative, as this total cost of ownership (TCO) analysis demonstrates.
The following analysis shows a cost savings of $91,400 of the tested ISE solution in this reference
architecture as compared to a similarly performant solution based on traditional enterprise storage arrays.
TCO versus Cost of Acquisition
The initial purchase price of any VDI system is just the beginning of the expenses that must be budgeted
and paid for. System requirements for rack space, power requirements and consumption, heat
generation, failure remediation, and more are all factors that increase the total cost of the solution when
calculated over the lifetime of the hardware (5 years).
Sizing Storage for Actual Performance Requirements
The Login VSI Medium User Workload test used to proof this reference architecture requires 6,000 to
8,000 write IOPS, while the Boot Storm test requires more than 25,000 write IOPS. Selecting storage
based on steady-state workload, rather than boot storm workload, almost always results in under-
performant VDI infrastructure and dissatisfied VDI users.
To illustrate this point: traditional enterprise storage vendors with products that utilize 10K RPM enterprise
SAS hard disk drives (the same drives used in the ISE 200 series arrays) typically rate their drives at 125
IOPS per drive. To deliver 8,000 IOPS for the Login VSI medium user workload test would require 64 of
these drives. To deliver Boot Storm (~25,000 IOPS) performance would require over 200 of these drives.
Figure 12 - Number of Disks Required for Login VSI Workloads
Purchasing a storage system based on steady-state user workload which, in this instance, results in an
under-resourced system by about 60% (136 additional drives) that are required for adequate performance
during boot storms.
And this problem is exacerbated in environments with a large number of users. These environments will
often see boot storm performance issues at the end of shifts, classes and work days, when users logoff
and the Gold Master boot image changes. If the business relies on shift workers (e.g., healthcare,
manufacturing, emergency response, customer support), or uses large numbers of VMs repetitively,
according to a schedule (as in academic and educational settings), boot storms will occur multiple times
each day.

Storage Density and Rack Space Consumption
The storage system that houses 200+ drives in our example above has significantly higher OPEX costs
than an equivalent pair of ISE 230s. In a typical configuration 200 drives require 9 disk shelves (24 drives
per shelf) and with associated storage controllers, will consume 22 rack units, as opposed to the six (6)
rack units consumed by the pair of ISEs.
In a colocation facility, where rack space costs are well-understood and passed on to end-use customers,
rack space can cost as much as $60 USD per rack unit, per month, making this 22 rack units for a
conventional 200 drive system almost $16,000 USD per year.
By contrast, the two ISEs from the configuration tested in this reference architecture deliver required
performance and required capacity while consuming just 6 rack units, for a yearly cost of $4,285 USD.
Figure 13 - Yearly Colocation Costs
To compare, the ISE solution savings over the 200 drive solution is $11,428 USD per year, or
approximately $60,000 USD over the five years a typical VDI implementation is in production.
Storage Energy Consumption
Power and cooling are significant expense lines for most data center operators.
The 200 drive solution described above typically consumes about 4.2 kW of energy per hour,
while the tested ISE 230 solution requires 1.4kW of direct power per hour.1
Indirect power (for HVAC, etc.) is assumed, for all storage arrays, to be equal to the direct power
requirements -- that doubles the energy requirement, increasing the energy consumed by the
conventional 200-drive solution to 8.38 kWh, and to 2.8 kWh for the tested ISE solution.
1
Direct power consumption for a conventional enterprise storage array is calculated as follows: 5.6 watts of power per disk drive;
230 watts of power per drive shelf; 500 watts of power per storage controller.

Figure 14 - Boot Storm Power Requirement
Of course, the price of a KwH of electricity varies widely, across the globe – gaps between high and low
spot-market prices are as high as $.40 USD per kWh. Normalizing our comparison using $.1423 per kWh
(commercial power rates in Boston, MA. USA, in November, 2013) yields the following comparative costs
for direct and indirect energy consumption over the 5 years of a typical VDI implementation.
Figure 15 - Boot Storm, Total Power Costs over 5 years
The ISE solution savings over the 200 drive solution is $34,302 USD over the five years a typical
VDI implementation is in production.

Storage Warranty Costs
Every ISE is backed by a five-year zero-cost warranty. By contrast, most enterprise storage arrays come
with a three-year warranty, and warranty costs for Years 4 and 5 can meet or exceed the initial purchase
price of the product: a cost designed to promote unnecessary upgrades well before the end of the
product’s useful life.
This TCO analysis demonstrates that, for a large VDI implementation, the ISE 200 series product is not
only a stellar price/performance option, but also that the ISE 200 series products will cost $90,000 less
than a conventional enterprise storage array to operate, over a five-year lifespan.
If comparative warranty costs are factored in to the TCO – as they should be – the cost advantage of the
ISE 200 series, in large-scale VDI environments, can approach $250,000 USD.
Summary
This document highlights a proven and tested architecture for a large-scale VDI deployment based on
Hewlett-Packard blade servers, Microsoft Windows Server 2012 R2 Hyper-V, Citrix XenDesktop 7 and the
X-IO Intelligent Storage Element (ISE) 230. This reference architecture readily supports the low-latency
storage performance and capacity demands of 750 medium-workload VDI users, both at boot time and in
normal steady-state operation. Additionally, the results presented in this document have been conducted
with the ISE 230 storage systems operating at 98% capacity utilization, proving the predictability of the
system’s performance for a real-world, representative VDI workload.
An ISE 230 base configuration for this reference architecture has an approximate acquisition street cost
of $90 per desktop, inclusive of a 5-year warranty. The total storage configuration in this reference
architecture has an approximate acquisition street cost of $200 per desktop, including two ISE 230
performance storage arrays for both write cache volumes and over 30 TB of user data volumes, fibre
channel networking and a 5-year hardware warranty for both systems.
For more information on how X-IO can solve some of the industry’s most prevalent performance,
predictability and reliability issues at unbeatable, industry leading costs, visit www.x-io.com.

Acknowledgements
Thin Client Computing
Thin Client Computing, a leading Citrix provider of Virtual Application, Desktop and Mobility solutions to
SMB and Enterprise clients, assisted materially in the configuration and testing done for this reference
architecture. For more information about Thin Client Computing visit www.thinclient.net
Citrix
Citrix Labs, Santa Clara provided the test facilities, and assisted materially in the configuration and testing
done for this reference architecture. For more information about Citrix, visit www.citrix.com
Login VSI
To simulate the closest possible production representation of a VDI solution architecture and workload,
Login VSI was used. For more information about Login VSI, visit www.loginvsi.com
Contact Information
Website:
http://www.x-io.com
Email:
fastforever@x-io.com
Get in touch with us:
Visit our website and chat
with us to get more information
http://www.x-io.com/contact
Talk to us
United States »
866.472.6764
International »
+1.925.298.6061
For more international numbers
please visit:
http://www.x-io.com/contact
9950 Federal Drive, Suite 100 | Colorado Springs, CO 80921 | U.S. >> 1.866.472.6764 | International. >> +1.925.298.6061 ; www.x-io.com
X-IO, X-IO Technologies, ISE and CADP are trademarks of Xiotech Corporation. Product names mentioned herein may be trademarks and/or
registered trademarks of their respective companies. © Xiotech Corporation. All rights reserved. Document Number RA-0001-20140227

Appendix A: ISE 230 Tested Configuration Details
Component Description
Storage array ISE 230
Controllers per ISE Two (2) highly redundant, active/active controllers
Drives per DataPac Twenty (20) 900 GB, 10K RPM 2.5” 6Gb/s SAS drives
DataPacs per ISE Two (2)
Total drives per ISE Forty (40)
Raw capacity per ISE 36 TB
Usable capacity after sparing 28.8 TB
Sparing 20% spare capacity
Performance with sparing User volumes are striped across both DataPacs providing
performance of 40 drives
Performance per ISE 13,500 IOPS
FC Connectivity per controller Four 8Gb FC ports
Management Ports per
controller
One 1Gb Ethernet port
Power per ISE 600 Watts (typical), 3.6 amps (208V)
Rack units per ISE 3U
Total drives Eighty (80)
Total DataPacs Four (4)
Total controllers Four (4)
Total useable capacity 57.6 TB usable pre-RAID
Total performance 27,000 IOPS
Total FC Connectivity Sixteen (16) 8Gb FC ports
As shown in Figure 2, only four FC ports (one FC port from each
controller) were used in the testing
Total management ports Four (4). Four management IP addresses were required
Total power 1.2 KW (typical), 7.2 amps (208V)
Total rack space consumption 6U

Citrix XenDesktop Reference Architecture for 750 users

More Related Content

What's hot

Similar to Citrix XenDesktop Reference Architecture for 750 users

Recently uploaded

Citrix XenDesktop Reference Architecture for 750 users