Dell Wyse Datacenter for VMware Horizon View … 2014 Updated to include 13g ... 3 Dell Wyse Datacenter for VMware Horizon View Reference Architecture ... Dell Wyse Datacenter for

A Dell Reference Architecture

Dell Wyse Datacenter for VMware Horizon View Reference Architecture

A Reference Architecture for the design, configuration and implementation of a VMware Horizon View environment.

Dell Wyse Solutions Engineering July 2015

2 Dell Wyse Datacenter for VMware Horizon View Reference Architecture | v.6.7

Revisions

Date Description

May 2014 Initial release (v.6.5)

November 2014 Updated to include 13g servers and increased VM density (v.6.6)

November 2014 Updated cloud client graphics and nomenclature (v.6.6.1)

July 2015 Updated density numbers for ESXi 6.0 and added PowerEdge C4130 (v.6.7)


THIS WHITE PAPER IS FOR INFORMATIONAL PURPOSES ONLY, AND MAY CONTAIN TYPOGRAPHICAL ERRORS AND

TECHNICAL INACCURACIES. THE CONTENT IS PROVIDED AS IS, WITHOUT EXPRESS OR IMPLIED WARRANTIES OF

ANY KIND.

© 2015 Dell Inc. All rights reserved. Reproduction of this material in any manner whatsoever without the express

written permission of Dell Inc. is strictly forbidden. For more information, contact Dell.

PRODUCT WARRANTIES APPLICABLE TO THE DELL PRODUCTS DESCRIBED IN THIS DOCUMENT MAY BE FOUND

AT: http://www.dell.com/learn/us/en/19/terms-of-sale-commercial-and-public-sector Performance of network

reference architectures discussed in this document may vary with differing deployment conditions, network loads, and

the like. Third party products may be included in reference architectures for the convenience of the reader. Inclusion

of such third party products does not necessarily constitute Dell’s recommendation of those products. Please consult

your Dell representative for additional information.

Trademarks used in this text:

Dell™, the Dell logo, Dell Boomi™, Dell Precision™ ,OptiPlex™, Latitude™, PowerEdge™, PowerVault™,

PowerConnect™, OpenManage™, EqualLogic™, Compellent™, KACE™, FlexAddress™, Force10™ and Vostro™ are

trademarks of Dell Inc. Other Dell trademarks may be used in this document. Cisco Nexus®, Cisco MDS®, Cisco NX-

0S®, and other Cisco Catalyst® are registered trademarks of Cisco System Inc. EMC VNX®, and EMC Unisphere® are

registered trademarks of EMC Corporation. Intel®, Pentium®, Xeon®, Core® and Celeron® are registered trademarks of

Intel Corporation in the U.S. and other countries. AMD® is a registered trademark and AMD Opteron™, AMD

Phenom™ and AMD Sempron™ are trademarks of Advanced Micro Devices, Inc. Microsoft®, Windows®, Windows

Server®, Internet Explorer®, MS-DOS®, Windows Vista® and Active Directory® are either trademarks or registered

trademarks of Microsoft Corporation in the United States and/or other countries. Red Hat® and Red Hat® Enterprise

Linux® are registered trademarks of Red Hat, Inc. in the United States and/or other countries. Novell® and SUSE® are

registered trademarks of Novell Inc. in the United States and other countries. Oracle® is a registered trademark of

Oracle Corporation and/or its affiliates. Citrix®, Xen®, XenServer® and XenMotion® are either registered trademarks or

trademarks of Citrix Systems, Inc. in the United States and/or other countries. VMware®, Virtual SMP®, vMotion®,

vCenter® and vSphere® are registered trademarks or trademarks of VMware, Inc. in the United States or other

countries. IBM® is a registered trademark of International Business Machines Corporation. Broadcom® and

NetXtreme® are registered trademarks of Broadcom Corporation. QLogic is a registered trademark of QLogic

Corporation. Other trademarks and trade names may be used in this document to refer to either the entities claiming

the marks and/or names or their products and are the property of their respective owners. Dell disclaims proprietary

interest in the marks and names of others.

http://www.dell.com/learn/us/en/19/terms-of-sale-commercial-and-public-sector


Table of contents Revisions ............................................................................................................................................................................................. 2

1 Introduction ................................................................................................................................................................................ 9

1.1 Purpose of this document ............................................................................................................................................. 9

1.2 Scope ................................................................................................................................................................................ 9

1.3 New in this release .......................................................................................................................................................... 9

2 Solution architecture overview ............................................................................................................................................. 10

2.1 Introduction ................................................................................................................................................................... 10

2.1.1 Physical architecture overview .................................................................................................................................... 11

2.1.2 Dell Wyse Datacenter – solution layers .................................................................................................................... 12

2.2 Local Tier 1 ..................................................................................................................................................................... 13

2.2.1 Local Tier 1 – 50 user combined pilot ...................................................................................................................... 13

2.2.2 Local Tier 1 – 50 user scale-ready pilot.................................................................................................................... 13

2.2.3 Local Tier 1 (iSCSI) ......................................................................................................................................................... 14

2.3 Shared Tier 1 – Rack ..................................................................................................................................................... 17

2.3.1 Shared Tier 1 – Rack – 555 users (iSCSI) .................................................................................................................. 17

2.3.2 Shared Tier 1 – Rack (iSCSI – EQL) ............................................................................................................................ 18

2.3.3 Shared Tier 1 – Rack – 1000 users (FC – CML) ...................................................................................................... 21

2.4 Shared Tier 1 – Blade ................................................................................................................................................... 24

2.4.1 Shared Tier 1 – Blade – 555 users (iSCSI – EQL) .................................................................................................... 24

2.4.2 Shared Tier 1 – Blade (iSCSI – EQL) .......................................................................................................................... 25

2.4.3 Shared Tier 1 – Blade (FC – CML) .............................................................................................................................. 28

3 Hardware components ........................................................................................................................................................... 31

3.1 Networking ..................................................................................................................................................................... 31

3.1.1 Force10 S55 (ToR switch) ............................................................................................................................................ 31

3.1.2 Force10 S60 (1Gb ToR switch) ................................................................................................................................... 32

3.1.3 Force10 S4810 (10Gb ToR switch) ............................................................................................................................. 33

3.1.4 Brocade 6510 (FC ToR switch) ................................................................................................................................... 35

3.1.5 PowerEdge M I/O Aggregator (10Gb blade interconnect) .................................................................................... 36

3.1.6 PowerConnect M6348 (1Gb blade interconnect) ................................................................................................... 36

3.1.7 Brocade M5424 (FC blade interconnect) ................................................................................................................. 37

3.2 Servers ............................................................................................................................................................................. 38


3.2.1 PowerEdge R730 ........................................................................................................................................................... 38

3.2.2 PowerEdge M620.......................................................................................................................................................... 38

3.3 Storage ............................................................................................................................................................................ 39

3.3.1 EqualLogic Tier 1 storage (iSCSI) ................................................................................................................................ 39

3.3.2 EqualLogic Tier 2 storage (iSCSI)............................................................................................................................... 40

3.3.3 Compellent storage (FC).............................................................................................................................................. 47

3.3.4 NAS .................................................................................................................................................................................. 50

3.4 Wyse Cloud Clients ....................................................................................................................................................... 51

3.4.1 Wyse 5020-P25 ............................................................................................................................................................. 51

3.4.2 Wyse 5012-D10DP ........................................................................................................................................................ 51

3.4.3 Wyse 7020-P45 ............................................................................................................................................................. 51

3.4.4 Wyse 7250-Z50D .......................................................................................................................................................... 52

3.4.5 Wyse 7290-Z90D7 ........................................................................................................................................................ 52

3.4.6 Wyse 7490-Z90Q8 ....................................................................................................................................................... 52

3.4.7 Dell Chromebook 11 ..................................................................................................................................................... 52

4 Software components ............................................................................................................................................................ 54

4.1 What's new in this release of Horizon View 6.0? .................................................................................................... 54

4.2 VMware Horizon View .................................................................................................................................................. 55

4.3 VDI hypervisor platform ............................................................................................................................................... 56

4.3.1 VMware vSphere 5 ........................................................................................................................................................ 56

5 Solution architecture ............................................................................................................................................................... 57

5.1 Compute server infrastructure ................................................................................................................................... 57

5.1.1 Local Tier 1 – Rack........................................................................................................................................................ 57

5.1.2 Shared Tier 1 – Rack ..................................................................................................................................................... 57

5.1.3 Shared Tier 1 – Blade ................................................................................................................................................... 58

5.2 Management server infrastructure ............................................................................................................................. 59

5.2.1 SQL databases .............................................................................................................................................................. 60

5.2.2 DNS ................................................................................................................................................................................. 60

5.3 Scaling guidance ........................................................................................................................................................... 61

5.3.1 Windows 7 – vSphere .................................................................................................................................................. 62

5.3.2 Windows 8 – vSphere .................................................................................................................................................. 63

5.3.3 Windows 8.1 – vSphere ............................................................................................................................................... 64


5.3.4 Windows 2008R2 – vSphere ...................................................................................................................................... 64

5.4 Storage architecture overview .................................................................................................................................... 65

5.4.1 Local Tier 1 storage ....................................................................................................................................................... 65

5.4.2 Shared Tier 1 storage .................................................................................................................................................... 65

5.4.3 Shared Tier 2 storage ................................................................................................................................................... 66

5.4.4 Storage networking – EqualLogic iSCSI ................................................................................................................... 66

5.4.5 Storage networking – Compellent Fibre Channel .................................................................................................. 68

5.5 Virtual networking ......................................................................................................................................................... 69

5.5.1 Local Tier 1 – Rack – iSCSI ......................................................................................................................................... 69

5.5.2 Shared Tier 1 – Rack – iSCSI ....................................................................................................................................... 71

5.5.3 Shared Tier 1 – Rack – Fibre Channel ....................................................................................................................... 74

5.5.4 Shared Tier 1 – Blade – iSCSI ..................................................................................................................................... 75

5.5.5 Shared Tier 1 – Blade – Fibre Channel ..................................................................................................................... 77

5.6 Solution high availability .............................................................................................................................................. 79

5.6.1 Compute layer HA (Local Tier 1) ............................................................................................................................... 80

5.6.2 vSphere HA (Shared Tier 1) .......................................................................................................................................... 81

5.6.3 Horizon View infrastructure protection .................................................................................................................... 81

5.6.4 Management server high availability ......................................................................................................................... 82

5.6.5 Horizon View VCS high availability ............................................................................................................................ 82

5.6.6 Windows File Services high availability ...................................................................................................................... 82

5.6.7 SQL Server high availability ......................................................................................................................................... 83

5.6.8 Load balancing .............................................................................................................................................................. 83

5.7 VMware Horizon View communication flow ........................................................................................................... 84

6 Customer-provided solution components ......................................................................................................................... 85

6.1 Customer-provided storage requirements .............................................................................................................. 85

6.2 Customer-provided switching requirements .......................................................................................................... 85

7 Solution performance and testing ........................................................................................................................................ 86

7.1 Load generation and monitoring ............................................................................................................................... 86

7.1.1 VMware View Planner................................................................................................................................................... 86

7.1.2 Login VSI – Login Consultants ................................................................................................................................... 86

7.1.3 Liquidware Labs Stratusphere UX ............................................................................................................................... 86

7.1.4 EqualLogic SAN HQ ...................................................................................................................................................... 87


7.1.5 VMware vCenter ............................................................................................................................................................ 87

7.2 Performance analysis methodology .......................................................................................................................... 88

7.2.1 Resource utilization ...................................................................................................................................................... 88

7.2.2 EUE tools information .................................................................................................................................................. 89

7.2.3 EUE real user information ............................................................................................................................................ 89

7.2.4 Dell Wyse Datacenter workloads and profiles ......................................................................................................... 89

7.2.5 Dell Wyse Datacenter profiles ................................................................................................................................... 90

7.2.6 Dell Wyse Datacenter workloads .............................................................................................................................. 90

7.2.7 Workloads running on shared graphics profile ....................................................................................................... 92

7.2.8 Workloads running on dedicated graphics profile .................................................................................................. 92

7.3 Testing and validation .................................................................................................................................................. 92

7.3.1 Testing process ............................................................................................................................................................. 92

7.4 VMware Horizon View test results ............................................................................................................................. 93

7.4.1 Configuration ................................................................................................................................................................. 93

7.4.2 ESXi 6.0/View 6.1 ........................................................................................................................................................... 95

7.5 Dell PowerEdge C4130 testing ................................................................................................................................. 103

7.5.1 Configuration ............................................................................................................................................................... 103

7.5.2 Test results ................................................................................................................................................................... 104

7.6 Dell EqualLogic PS6210XS testing with VMware Horizon View ......................................................................... 110

7.6.1 Overview ....................................................................................................................................................................... 110

7.6.2 Compute resources ..................................................................................................................................................... 111

7.6.3 Network resources ...................................................................................................................................................... 111

7.6.4 iSCSI SAN configuration overview ........................................................................................................................... 112

7.6.5 Test objectives: ............................................................................................................................................................ 112

7.6.6 Test criteria/thresholds: ............................................................................................................................................. 113

7.6.7 Boot storm I/O ............................................................................................................................................................ 113

7.6.8 Login storm I/O ........................................................................................................................................................... 114

7.6.9 Steady state I/O ........................................................................................................................................................... 116

7.6.10 Server host performance ...................................................................................................................................... 118

7.6.11 Summary ....................................................................................................................................................................... 119

Acknowledgements ...................................................................................................................................................................... 121

About the authors ......................................................................................................................................................................... 121



1 Introduction

1.1 Purpose of this document This document describes:

Dell Wyse Datacenter for VMware Horizon View Reference Architecture scaling from 50 to

50,000+ virtual desktop infrastructure (VDI) users.

Solution options encompass a combination of solution models including local disks, iSCSI or Fibre

Channel based storage options.

This document addresses the architecture design, configuration and implementation considerations for

the key components of the architecture required to deliver virtual desktops via VMware Horizon View on

VMware vSphere 5.

1.2 Scope Relative to delivering the virtual desktop environment, the objectives of this document are to:

Define the detailed technical design for the solution.

Define the hardware requirements to support the design.

Define the design constraints which are relevant to the design.

Define relevant risks, issues, assumptions and concessions – referencing existing ones where

possible.

Provide a breakdown of the design into key elements such that the reader receives an incremental

or modular explanation of the design.

Provide solution scaling and component selection guidance.

1.3 New in this release RDS based desktop and Remote App support - http://dell.to/QRqAud

View 6 Cloud POD Architecture - http://dell.to/1gOGrB5

See the attached hyperlinks for focused white papers on each of the above topics.

http://dell.to/QRqAud

http://dell.to/QRqAud

http://dell.to/1gOGrB5


2 Solution architecture overview

2.1 Introduction The Dell Wyse Datacenter Solution leverages a core set of hardware and software components consisting

of 4 primary layers:

Networking Layer

Compute Server Layer

Management Server Layer

Storage Layer

These components have been integrated and tested to provide the optimal balance of high performance

and lowest cost per user. Additionally, the Dell Wyse Datacenter Solution includes an approved extended

list of optional components in the same categories. These components give IT departments the flexibility

to custom tailor the solution for environments with unique virtual desktop infrastructure (VDI) feature,

scale or performance needs. The Dell Wyse Datacenter stack is designed to be a cost effective starting

point for IT departments looking to migrate to a fully virtualized desktop environment slowly. This

approach allows you to grow the investment and commitment as needed or as your IT staff becomes

more comfortable with VDI technologies.


2.1.1 Physical architecture overview The core Dell Wyse Datacenter architecture consists of two models: Local Tier 1 and Shared Tier 1. Tier 1

in the Dell Wyse Datacenter context defines from which disk source the VDI sessions execute. Local Tier 1

includes rack servers only while Shared Tier 1 can include rack or blade servers due to the usage of shared

Tier 1 storage. Tier 2 storage is present in both solution architectures and, while having a reduced

performance requirement, is utilized for user profile/data and Management virtual machine (VM)

execution. Management VM execution occurs using Tier 2 storage for all solution models. Dell Wyse

Datacenter is a 100% virtualized solution architecture.

In the Shared Tier 1 solution model, an additional high-performance shared storage array is added to

handle the execution of the VDI sessions. All compute and management layer hosts in this model are

diskless.

User DataMgmt Disk

MGMT Server

CPU RAM

T2 Shared Storage

Mgmt VMs

VDI VMs

Compute Server

CPU RAMVDI Disk

Local Tier 1

VDI DiskUser Data

MGMT Server

CPU RAM

T1 Shared Storage

Mgmt VMs VDI VMs

Compute Server

CPU RAM

T2 Shared Storage

Mgmt Disk

Shared Tier 1


2.1.2 Dell Wyse Datacenter – solution layers Only a single high performance Force10 48-port switch is required to get started in the network layer. This

switch will host all solution traffic consisting of 1Gb iSCSI and LAN sources for smaller stacks. Above 1000

users we recommend that LAN and iSCSI traffic be separated into discrete switching fabrics. Additional

switches can be added and stacked as required to provide High Availability for the Network layer.

The compute layer consists of the server resources responsible for hosting the Horizon View user

sessions, hosted via the VMware vSphere hypervisor, local or shared tier 1 solution models (local Tier 1

pictured below).

VDI management components are dedicated to their own layer so as to not negatively impact the user

sessions running in the compute layer. This physical separation of resources provides clean, linear and

predictable scaling without the need to reconfigure or move resources within the solution as you grow.

The management layer will host all the VMs necessary to support the VDI infrastructure.

The storage layer consists of options provided by EqualLogic for iSCSI and Compellent arrays for Fibre

Channel to suit your Tier 1 and Tier 2 scaling and capacity needs.


2.2 Local Tier 1

2.2.1 Local Tier 1 – 50 user combined pilot For a very small deployment or pilot effort to familiarize you with the solution architecture, we offer a 50

user combined pilot solution. This architecture is non-distributed with all VDI, Management and storage

functions on a single host running vSphere. If additional scaling is desired, you can grow into a larger

distributed architecture seamlessly with no loss on initial investment.

2.2.2 Local Tier 1 – 50 user scale-ready pilot In addition to the 50 user combined offering we also offer a scale ready version that includes Tier 2

storage. The basic architecture is the same but customers looking to scale out quickly will benefit by

building out into Tier 2 initially.


2.2.3 Local Tier 1 (iSCSI) The Local Tier 1 solution model provides a scalable rack-based configuration that hosts user VDI sessions

on local disk in the compute layer.


2.2.3.1 Local Tier 1 – network architecture (iSCSI) In the local tier 1 architecture, a single Force10 switch can be shared among all network connections for

both management and compute, up to 1000 users. Over 1000 users Dell Wyse Solutions Engineering

recommends separating the network fabrics to isolate iSCSI and LAN traffic as well as making each switch

stack redundant. Only the management servers connect to iSCSI storage in this model. All Top of Rack

(ToR) traffic has been designed to be layer 2 (switched locally), with all layer 3 (routable) VLANs trunked

from a core or distribution switch. The following diagrams illustrate the logical data flow in relation to the

core switch.

DR

AC

VL

AN

Mg

mt

VL

AN

iSCSI

Mgmt hosts

Compute hosts

Core switch

Tru

nk

SAN

VD

I VLA

N

ToR switches

vMo

tio

n V

LA

N


2.2.3.2 Local Tier 1 cabling diagram for high availability (HA) (Rack – HA)

2.2.3.3 Local Tier 1 rack scaling guidance (iSCSI)

Local Tier 1 Hardware Scaling (iSCSI)

User Scale ToR LAN ToR 1Gb iSCSI EQL T2 EQL NAS

1-1000 S55 PS4100E -

1-1000 (HA) S55 S55 PS4100E FS7600

1-3000 S55 S55 PS6100E FS7600

3000-6000 S55 S55 PS6500E FS7600

6000+ S60 S60 PS6500E FS7600

SAN

LAN

S55/S60

S55/S60


2.3 Shared Tier 1 – Rack

2.3.1 Shared Tier 1 – Rack – 555 users (iSCSI) For proofs of concept (POCs) or small deployments, Tier 1 and Tier 2 can be combined on a single

PS6210XS storage array. Above 555 users, a separate array needs to be used for Tier 2.


2.3.2 Shared Tier 1 – Rack (iSCSI – EQL) For 555 or more users on EqualLogic (EQL), the Storage layers are separated into discrete arrays. The

drawing below depicts a 3000 user build where the network fabrics are separated for LAN and iSCSI traffic.

Additional PS6210XS arrays are added for Tier 1 as the user count scales, just as the Tier 2 array models

change also based on scale. The PS4110E, PS6210E and PS6510E are 10Gb Tier 2 array options. NAS is

recommended above 1000 users to provide HA for file services.


2.3.2.1 Shared Tier 1 – Rack – network architecture (iSCSI) In the Shared Tier 1 architecture for rack servers, both management and compute servers connect to

shared storage in this model. All ToR traffic has designed to be layer 2 (switched locally), with all layer 3

(routable) VLANs routed through a core or distribution switch. The following diagrams illustrate the server

NIC to ToR switch connections, vSwitch assignments, as well as logical VLAN flow in relation to the core

switch.

DR

AC

VL

AN

Mg

mt

VL

AN

iSCSI

Mgmt hosts

Compute hosts

Core switch

ToR switches

Tru

nk

SAN

vMo

tio

n V

LA

N

VD

I VLA

N


2.3.2.2 Shared Tier 1 – Rack – Cabling diagram (Rack – EQL)

2.3.2.3 Shared Tier 1 – Rack – Scaling guidance (iSCSI)

Shared Tier 1 Hardware Scaling (Rack – iSCSI)

User Scale ToR LAN ToR 1Gb iSCSI EQL T1 EQL T2 EQL NAS

1-500 S55 S4810 PS6210XS - -

500-1000 S55 S4810 PS6210XS PS4110E -

1-1000 (HA) S55 S4810 PS6210XS PS4110E NX3300

1-3000 S55 S4810 PS6210XS PS6210E NX3300

3000-6000 S55 S4810 PS6210XS PS6510E NX3300

6000+ S60 S4810 PS6210XS PS6510E NX3300

SAN

LAN

S4810S55/S60


2.3.3 Shared Tier 1 – Rack – 1000 users (FC – CML) Utilizing Compellent (CML) storage for Shared Tier 1 provides a Fibre Channel solution where Tier 1 and

Tier 2 can optionally be combined in a single array. Tier 2 functions (user data + management VMs) can be

removed from the array if the customer has another tier 2 solution in place or a tier 2 Compellent array

can be used. Scaling this solution is very linear by predictably adding Compellent arrays for every 2000

basic users, on average. The image below depicts a 1000 user array. For 2000 users, 96 total disks in 4

shelves are required. Please see section 3.3.3 for more information.


2.3.3.1 Shared Tier 1 – Rack – Network architecture (FC) In the Shared Tier 1 architecture for rack servers using Fibre Channel (FC), a separate switching

infrastructure is required for FC. Management and compute servers will both connect to shared storage

using FC. Both management and compute servers connect to all network VLANs in this model. All ToR

traffic has designed to be layer 2 (switched locally), with all layer 3 (routable) VLANs routed through a core

or distribution switch. The following diagrams illustrate the server NIC to ToR switch connections, vSwitch

assignments, as well as logical VLAN flow in relation to the core switch.

DR

AC

VL

AN

Mg

mt

VL

AN

FC

Mgmt hosts

Compute hosts

Core switch

Tru

nk

SAN

VD

I VL

AN

ToR Ethernet switch

vM

oti

on

VL

AN

FC switch


2.3.3.2 Shared Tier 1 – Rack – Cabling diagram (Rack – CML)

2.3.3.3 Shared Tier 1 – Rack – Scaling guidance (FC)

Shared Tier 1 Hardware Scaling (Rack – FC)

User Scale LAN Network FC Network CML T1 CML T2 CML NAS

1-1000 S55 6510 SC8000, 15K SAS - -

1-1000 (HA) S55 6510 SC8000, 15K SAS SC8000, NL-SAS FS8600

1000-6000 S55 6510 SC8000, 15K SAS SC8000, NL-SAS FS8600

6000+ S60 6510 SC8000, 15K SAS SC8000, NL-SAS FS8600

SAN

LAN

6510S55/S60


2.4 Shared Tier 1 – Blade

2.4.1 Shared Tier 1 – Blade – 555 users (iSCSI – EQL) As is the case in the Shared Tier 1 model using rack servers, blades can also be used in a 500 user bundle

by combing Tier 1 and Tier 2 on a single PS6210XS array. Above 555 users, separate Tier 1 and Tier 2

storage into discrete arrays.


2.4.2 Shared Tier 1 – Blade (iSCSI – EQL) Above 1000 users the storage tiers need to be separated to maximize the performance of the PS6210XS

for VDI sessions. At this scale we also separate LAN from iSCSI switching. Optionally, load balancing and

NAS can be added for HA. The drawing below depicts a 3000 user solution.


2.4.2.1 Shared Tier 1 – Blade – Network architecture (iSCSI) In the Shared Tier 1 architecture for blades, only iSCSI is switched through a ToR switch. There is no need

to switch LAN ToR since the M6348 in the chassis supports LAN to the blades and can be uplinked to the

core or distribution layers directly. The M6348 has 16 external ports per switch that can be optionally used

for DRAC/IPMI traffic. For greater redundancy, a ToR switch used to support DRAC/IPMI can be used

outside of the chassis. Both Management and Compute servers connect to all VLANs in this model. The

following diagram illustrates the server NIC to ToR switch connections, vSwitch assignments, as well as

logical VLAN flow in relation to the core switch.

DR

AC

VL

AN

Mg

mt

VL

AN

iSCSI

Mgmt hosts

Compute hosts

Core switch

ToR switch

Tru

nk

SAN

vMo

tio

n V

LA

N

VD

I VLA

N


2.4.2.2 Shared Tier 1 – Blade – Cabling diagram (Blade – EQL)

2.4.2.3 Shared Tier 1 – Blade – Scaling guidance (iSCSI)

Shared Tier 1 Hardware Scaling (Blade – iSCSI)

User Scale Blade LAN Blade iSCSI ToR 10Gb iSCSI

EQL T1 EQL T2 EQL NAS

1-500 M6348 IOA S4810 PS6210XS - -

500-1000 M6348 IOA S4810 PS6210XS PS4110E -

1-1000 (HA) M6348 IOA S4810 PS6210XS PS4110E NX3300

1-3000 M6348 IOA S4810 PS6210XS PS6110E NX3300

3000-6000 M6348 IOA S4810 PS6210XS PS6510E NX3300

6000+ M6348 IOA S4810 PS6210XS PS6510E NX3300

S4810 Stack

Core

10Gb SAN

10Gb LAN

Stacking


2.4.3 Shared Tier 1 – Blade (FC – CML) Fibre Channel is again an option in Shared Tier 1 using blades. There are a few key differences using FC

with blades instead of iSCSI: Blade chassis interconnects, FC HBAs in the servers and FC IO cards in the

Compellent arrays. ToR FC switching is optional if a suitable FC infrastructure is already in place. The

image below depicts a 4000 user stack.


2.4.3.1 Shared Tier 1 – Blade – Network architecture (FC)

DR

AC

VL

AN

Mg

mt

VL

AN

FC

Mgmt hosts

Compute hosts

Core switch

FC switch

Tru

nk

SAN

vM

oti

on

VL

AN

VD

I VL

AN


2.4.3.2 Shared Tier 1 – Blade – Cabling diagram (Blade – CML)

2.4.3.3 Shared Tier 1 – Blade – Scaling guidance (FC)

Shared Tier 1 Hardware Scaling (Blade – FC)

User Scale Blade LAN Blade FC ToR FC CML T1 CML T2 CML NAS

1-500 IOA 5424 6510 SC8000, 15K SAS - -

500-100 IOA 5424 6510 SC8000, 15K SAS - -

1-1000 (HA) IOA 5424 6510 SC8000, 15K SAS SC8000, NL SAS FS8600

1000-6000 IOA 5424 6510 SC8000, 15K SAS SC8000, NL SAS FS8600

6000+ IOA 5424 6510 SC8000, 15K SAS SC8000, NL SAS FS8600

6510FC fabric A

Core

6510FC fabric B

FC SAN

10Gb LAN

Stacking


3 Hardware components

3.1 Networking The following sections contain the core network components for the Dell Wyse Datacenter solutions.

General uplink cabling guidance to consider in all cases is that Twinax is very cost effective for short 10Gb

runs and for longer runs it is best to use fiber with SFPs.

3.1.1 Force10 S55 (ToR switch) The Dell Force10 S-Series S55 1/10 GbE Top-of-Rack (ToR) switch is optimized for lowering operational

costs while increasing scalability and improving manageability at the network edge. Optimized for high-

performance data center applications, the S55 is recommended for Dell Wyse Datacenter deployments of

6000 users or less and leverages a non-blocking architecture that delivers line-rate, low-latency L2 and L3

switching to eliminate network bottlenecks. The high-density S55 design provides 48 GbE access ports

with up to four modular 10 GbE uplinks in just 1-RU to conserve valuable rack space. The S55 incorporates

multiple architectural features that optimize data center network efficiency and reliability, including IO

panel to PSU airflow or PSU to IO panel airflow for hot/cold aisle environments and redundant, hot-

swappable power supplies and fans. A “scale-as-you-grow” ToR solution that is simple to deploy and

manage, up to 8 S55 switches can be stacked to create a single logical switch by utilizing Dell Force10’s

stacking technology and high-speed stacking modules.

Model Features Options Uses

Force10 S55 44 x BaseT (10/100/1000) + 4 x SFP

Redundant PSUs

4 x 1Gb SFP (Cu or fiber)

12 or 24Gb stacking port (up to 8 switches)

2 x slots for 10Gb uplink or stacking modules

ToR switch for LAN and iSCSI in Local Tier 1 solution

Guidance:


10Gb uplinks to a core or distribution switch are the preferred design choice using the rear 10Gb

uplink modules. If 10Gb to a core or distribution switch is unavailable the front 4 x 1Gb SFP ports

can be used.

The front 4 SFP ports can support copper cabling and can be upgraded to optical if a longer run is

needed.

For more information on the S55 switch and Dell Force10 networking, please visit:

http://www.dell.com/us/enterprise/p/force10-s55/pd

3.1.1.1 Force10 S55 stacking The Top of Rack switches in the Network layer can be optionally stacked with additional switches, if

greater port count or redundancy is desired. Each switch will need a stacking module plugged into a rear

bay and connected with a stacking cable. The best practice for switch stacks greater than 2 is to cable in a

ring configuration with the last switch in the stack cabled back to the first. Uplinks need to be configured

on all switches in the stack back to the core to provide redundancy and failure protection.

3.1.2 Force10 S60 (1Gb ToR switch) The Dell Force10 S-Series S60 is a high-performance 1/10 GbE access switch optimized for lowering

operational costs at the network edge and is recommended for Dell Wyse Datacenter deployments over

6000 users. The S60 answers the key challenges related to network congestion in data center ToR (Top-

of-Rack) and service provider aggregation deployments. As the use of large data burst applications and

services continue to increase, huge spikes in network traffic that can cause network congestion and

packet loss also become more common. The S60 is equipped with the industry’s largest packet buffer

(1.25 GB), enabling it to deliver lower application latency and maintain predictable network performance

even when faced with significant spikes in network traffic. Providing 48 line-rate GbE ports and up to four

optional 10 GbE uplinks in just 1-RU, the S60 conserves valuable rack space. Further, the S60 design

delivers unmatched configuration flexibility, high reliability and power and cooling efficiency to reduce

costs.


Force10 S60 44 x BaseT (10/100/1000) + 4 x SFP

High performance

High scalability

Redundant PSUs

4 x 1Gb SFP (Cu or fiber)

12 or 24Gb stacking ports (up to 12 switches)

2 x slots for 10Gb uplink or stacking modules

Higher scale ToR switch for LAN in Local + Shared Tier 1 and iSCSI in Local Tier 1 solution



Guidance:

10Gb uplinks to a core or distribution switch is the preferred design choice using the rear 10Gb

uplink modules. If 10Gb to a core or distribution switch is unavailable the front 4 x 1Gb SFP ports

can be used.


needed.

The S60 is appropriate for use in solutions scaling higher than 6000 users.



3.1.2.1 S60 stacking The S60 switch can be optionally stacked with 2 or more switches, if greater port count or redundancy is

desired. Each switch will need a stacking module plugged into a rear bay and connected with a stacking

cable. The best practice for switch stacks greater than 2 is to cable in a ring configuration with the last

switch in the stack cabled back to the first. Uplinks need to be configured on all switches in the stack back

to the core to provide redundancy and failure protection.

3.1.3 Force10 S4810 (10Gb ToR switch) The Dell Force10 S-Series S4810 is an ultra-low latency 10/40 GbE Top-of-Rack (ToR) switch purpose-

built for applications in high-performance data center and computing environments. Leveraging a non-

blocking, cut-through switching architecture, the S4810 delivers line-rate L2 and L3 forwarding capacity

with ultra-low latency to maximize network performance. The compact S4810 design provides industry-

leading density of 48 dual-speed 1/10 GbE (SFP+) ports as well as four 40 GbE QSFP+ uplinks to conserve

valuable rack space and simplify the migration to 40 Gbps in the data center core (Each 40 GbE QSFP+



uplink can support four 10 GbE ports with a breakout cable). Priority-based Flow Control (PFC), Data

Center Bridge Exchange (DCBX), Enhance Transmission Selection (ETS), coupled with ultra-low latency

and line rate throughput, make the S4810 ideally suited for iSCSI storage, FCoE Transit & DCB

environments.


Force10 S4810

48 x SFP+ (1Gb/10Gb) + 4 x QSFP+ (40Gb)

Redundant PSUs

Single-mode/multi-mode optics, Twinax, QSFP+ breakout cables

Stack up to 6 switches with SFP or QSFP (2 with VLT)

ToR switch for iSCSI in Shared Tier 1 solution

Guidance:

The 40Gb QSFP+ ports can be split into 4 x 10Gb ports using breakout cables for stand-alone

units, if necessary. This is not supported in stacked configurations.

10Gb or 40Gb uplinks to a core or distribution switch is the preferred design choice.


needed.

The S60 is appropriate for use in solutions scaling higher than 6000 users.



3.1.3.1 S4810 stacking The S4810 switch can be optionally stacked up to 6 switches or configured to use Virtual Link Trunking

(VLT) up to 2 switches. Stacking is supported on either SFP or QSFP ports as long as that port is configured

for stacking. The best practice for switch stacks greater than 2 is to cable in a ring configuration with the

last switch in the stack cabled back to the first. Uplinks need to be configured on all switches in the stack

back to the core to provide redundancy and failure protection.



3.1.4 Brocade 6510 (FC ToR switch) The Brocade 6510 Switch meets the demands of hyper-scale, private cloud storage environments by

delivering market-leading speeds up to 16 Gbps Fibre Channel technology and capabilities that support

highly virtualized environments. Designed to enable maximum flexibility and investment protection, the

Brocade 6510 is configurable in 24, 36, or 48 ports and supports 2, 4, 8, or 16 Gbps speeds in an efficiently

designed 1U package. It also provides a simplified deployment process and a point-and-click user

interface—making it both powerful and easy to use. The Brocade 6510 offers low-cost access to industry-

leading Storage Area Network (SAN) technology while providing “pay-as-you-grow” scalability to meet the

needs of an evolving storage environment.


Brocade 6510 48 x 2/4/8/16Gb FC

Additional FlexIO module (optional)

Up to 24 total ports (internal + external)

Ports on demand from 24, 36 and 48.

FC ToR switch for all solutions. Optional for blades.

Guidance:

The 6510 FC switch can be licensed to light the number of ports required for the deployment. If

only 24 or fewer ports are required for a given implementation, then only those need to be

licensed.

Up to 239 Brocade switches can be used in a single FC fabric.

For more information on the Brocade 6510 switch, please visit:

http://www.dell.com/us/enterprise/p/brocade-6510/pd

48 x Auto-sensing ports

http://www.dell.com/us/enterprise/p/brocade-6510/pd


3.1.5 PowerEdge M I/O Aggregator (10Gb blade interconnect)


PowerEdge M I/O Aggregator (IOA)

Up to 32 x 10Gb ports + 4 x external SFP+

2 x line rate fixed QSFP+ ports

2 x FlexIO module bays

2-port QSFP+ modules in 4x10Gb mode

4-port SFP+ 10Gb module

4-port 10GBaseT copper module (one per IOA)

Stacking available only with Active System Manager

Blade switch for iSCSI in Shared Tier 1 blade solution

Guidance:

10Gb uplinks to a ToR switch are the preferred design choice using Twinax or optical cabling for

longer runs.

If copper-based uplinks are necessary, additional FlexIO modules can be used.

For more information on the Dell IOA switch, please visit:

http://www.dell.com/us/business/p/poweredge-m-io-aggregator/pd

3.1.6 PowerConnect M6348 (1Gb blade interconnect)


PowerConnect M6348

32 x internal (1Gb)

16 x external Base-T

2 x 10Gb SFP+

2 x 16Gb stacking/CX4 ports

Stack up to 12 switches Blade switch for LAN traffic in Shared Tier 1 blade solution

Guidance:

10Gb uplinks to a core or distribution switch are the preferred design choice using Twinax or

optical cabling via the SFP+ ports.

16 x external 1Gb ports can be used for Management ports, DRACs, etc.

FlexIO slots to provide additional copper or SFP+ ports

2 x QSFP+ ports (4x10Gb)

XG

2

XG

1

XG

3

CO

NS

OL

E

33

34

LK

M6

34

8

47

48

Ac

t

LK

Ac

t

LK

Ac

t

XG

4

2 x 1Gb/10Gb SFP+ uplink ports16 x 1Gb Base-T ports

2 x 16Gb Stacking/ CX4 ports

http://www.dell.com/us/business/p/poweredge-m-io-aggregator/pd


Stack up to 12 switches using stacking ports.

3.1.7 Brocade M5424 (FC blade interconnect) The Brocade M5424 switch and the Dell PowerEdge M1000e blade enclosure provide robust solutions for

Fibre Channel SAN deployments. Not only does this offering help simplify and reduce the amount of SAN

hardware components required for a deployment, but it also maintains the scalability, performance,

interoperability and management of traditional SAN environments. The M5424 can easily integrate FC

technology into new or existing SAN environments using the PowerEdge M1000e blade enclosure. The

Brocade M5424 is a flexible platform that delivers advanced functionality, performance, manageability,

scalability with up to 16 internal fabric ports and up to 8 2GB/4GB/8GB auto-sensing uplinks and is ideal

for larger storage area networks. Integration of SAN switching capabilities with the M5424 also helps to

reduce complexity and increase SAN manageability.


Brocade M5424

Up to 8 x 2/4/8Gb auto-sensing uplinks

16 x internal fabric ports

Ports on demand from 12 to 24.

Blade switch for FC in Shared Tier 1 model.

Guidance:

12-port model includes 2 x 8Gb transceivers, 24-port models include 4 or 8 transceivers.

Up to 239 Brocade switches can be used in a single FC fabric.

3.1.7.1 QLogic QME2572 host bus adapter

The QLogic QME2572 is a dual-channel 8Gb/s Fibre Channel host bus

adapter (HBA) designed for use in PowerEdge M1000e blade servers.

Doubling the throughput enables higher levels of server consolidation

and reduces data-migration/backup windows. It also improves

performance and ensures reduced response time for mission-critical

and next generation killer applications. Optimized for virtualization,

power, security and management, as well as reliability, availability and

serviceability (RAS), the QME2572 delivers 200,000 IOs per second

(IOPS).


3.1.7.2 QLogic QLE2562 HBA The QLE2562 is a PCI Express, dual port, Fibre Channel HBA. The QLE2562 is

part of the QLE2500 HBA product family that offers next generation 8 Gb FC

technology, meeting the business requirements of the enterprise data center.

Features of this HBA includes throughput of 3200 MBps (full-duplex), 200,000

initiator and target I/Os per second (IOPS) per port and StarPower technology-

based dynamic and adaptive power management. Benefits include optimizations

for virtualization, power, reliability, availability and serviceability (RAS) and

security.

3.2 Servers

3.2.1 PowerEdge R730

Pow erEdge R720

The rack server platform for the Dell Wyse Datacenter solution is the best-in-class Dell PowerEdge R730.

This dual socket CPU platform runs the fastest Intel Xeon E5-2600 v3 family of processors, can host up to

768GB RAM and supports up to 16 2.5” SAS disks. The Dell PowerEdge R730 offers uncompromising

performance and scalability in a 2U form factor. For more information, please visit:

http://www.dell.com/us/business/p/poweredge-r730/pd

3.2.2 PowerEdge M620

http://www.dell.com/us/business/p/poweredge-r730/pd


The blade server platform for the Dell Wyse Datacenter solution is the PowerEdge M620. This half-height

blade server is a feature-rich, dual-processor platform that offers a blend of density, performance,

efficiency and scalability. The M620 offers remarkable computational density, scaling up to 24 cores, 2

socket Intel Xeon processors and 24 DIMMs (768GB RAM) of DDR3 memory in an extremely compact

half-height blade form factor. This server platform is currently offered in both the PowerEdge M1000e

blade enclosure and VRTX shared infrastructure platform. For more information, please visit:

http://www.dell.com/us/business/p/poweredge-m620/pd

3.3 Storage

3.3.1 EqualLogic Tier 1 storage (iSCSI)

3.3.1.1 PS6210XS Implement both high-speed, low-latency solid-state disk (SSD) technology and high-capacity HDDs from

a single chassis. The PS6210XS 10GbE iSCSI array is a Dell Fluid Data solution with a virtualized scale-out

architecture that delivers enhanced storage performance and reliability that is easy to manage and scale

for future needs. For more information please visit: http://www.dell.com/us/business/p/equallogic-

ps6210-series/pd


EqualLogic PS6210XS

24-drive hybrid array (SSD + 10K SAS)

Dual HA controllers

Snaps/clones

Asynchronous replication

SAN HQ,

4 x 10Gb interfaces per controller (2 x SFP + 2 x 10GBT)

13 TB – 7 x 400GB SSD + 17 x 600GB 10K SAS

Tier 1 array for Shared Tier 1 solution model (10Gb – iSCSI)

26 TB – 7 x 800GB SSD + 17 x 1.2TB 10K SAS

Tier 1 array for Shared Tier 1 solution model requiring greater per user capacity (10Gb – iSCSI)

http://www.dell.com/us/business/p/poweredge-m620/pd

http://www.dell.com/us/business/p/equallogic-ps6210-series/pd

http://www.dell.com/us/business/p/equallogic-ps6210-series/pd


3.3.2 EqualLogic Tier 2 storage (iSCSI) The following arrays can be used for management VM storage and user data, depending on the scale of

the deployment. Please refer to the hardware tables in section 2 or the “Uses” column of each array below.

For more information on Dell EqualLogic offerings, please visit: http://www.dellstorage.com/equallogic/

3.3.2.1 PS4100E


EqualLogic PS4100E

12 drive bays (NL-SAS/ 7200 RPM)

Dual HA controllers

Snaps/clones


SAN HQ

1Gb

12TB – 12 x 1TB HDDs Tier 2 array for 1000 users or less in Local Tier 1 solution model (1Gb – iSCSI) 24TB – 12 x 2TB HDDs

36TB – 12 x 3TB HDDs

7 x SSD + 17 x 10K SAS

10Gb Ethernet ports

http://www.dellstorage.com/equallogic/


MANAGEMENT

SERIAL PORT

SERIAL PORT

MANAGEMENT

STANDBY

ON/OFF

ACT

ERR

PWR

ACT

ERR

PWR

STANDBY

ON/OFF

ETHERNET 1

ETHERNET 1

ETHERNET 0

CONTROL MODULE 12

ETHERNET 0

CONTROL MODULE 12

0

4

8

1

5

9

2

6

10

3

7

11

Hard Drives

1Gb Ethernet ports Mgmt ports

12 x NL SAS drives


3.3.2.2 PS4110E


EqualLogic PS4110E


Dual HA controllers

Snaps/clones


SAN HQ

10Gb

12TB – 12 x 1TB HDDs Tier 2 array for 1000 users or less in Shared Tier 1 solution model (10Gb – iSCSI) 24TB – 12 x 2TB HDDs



3.3.2.3 PS6100E


EqualLogic PS6100E


Dual HA controllers

Snaps/clones


SAN HQ

1Gb

4U chassis

24TB – 24 x 1TB HDDs Tier 2 array for up to 1500 users, per array, in local Tier 1 solution model (1Gb) 48TB – 24 x 2TB HDDs




3.3.2.4 PS6210E


EqualLogic PS6210E


Dual HA controllers

Snaps/clones


SAN HQ

10Gb

4U chassis

24TB – 24 x 1TB HDDs Tier 2 array for up to 1500 users, per array, in shared Tier 1 solution model (10Gb) 48TB – 24 x 2TB HDDs



24 x 7.2K NL-SAS drives

10Gb Ethernet ports Mgmt ports


3.3.2.5 PS6500E


EqualLogic PS6500E

48 drive SATA (NL-SAS)

Dual HA controllers

Snaps/clones


SAN HQ

1Gb

48TB – 48 x 1TB HDDs Tier 2 array for Local Tier 1 solution model (1Gb – iSCSI)




3.3.2.6 PS6510E


EqualLogic PS6510E

48 drive SATA (NL-SAS)

Dual HA controllers

Snaps/clones


SAN HQ

10Gb

48TB – 48 x 1TB HDDs Tier 2 array for Shared Tier 1 solution model (10Gb – iSCSI)




3.3.2.7 EqualLogic configuration Each tier of EqualLogic storage is to be managed as a separate pool or group to isolate specific workloads.

Manage shared Tier 1 arrays used for hosting VDI sessions together, while managing shared Tier 2 arrays

used for hosting Management server role VMs and user data together.

3.3.3 Compellent storage (FC)

Dell Wyse Solutions Engineering recommends that all Compellent storage arrays be implemented using 2

controllers in an HA cluster. Fibre Channel is the preferred storage protocol for use with this array, but

Compellent is fully capable of supporting iSCSI as well. Key Storage Center applications used strategically

to provide increased performance include:

Fast Track – Dynamic placement of most frequently accessed data

blocks on the faster outer tracks of each spinning disk. Lesser active

data blocks remain on the inner tracks. Fast track is well-

complimented when used in conjunction with Thin Provisioning.

Data Instant Replay – Provides continuous data protection using

snapshots called Replays. Once the base of a volume has been captured, only incremental

changes are then captured going forward. This allows for a high number of Replays to be

scheduled over short intervals, if desired, to provide maximum protection.

3.3.3.1 Compellent Tier 1 Compellent Tier 1 storage consists of a standard dual controller configuration and scales upward by

adding disks/shelves and additional discrete arrays. A single pair of SC8000 controllers will support Tier 1

and Tier 2 for 2000 knowledge worker users, as depicted below, utilizing all 15K SAS disks. If Tier 2 is to be

separated then an additional 30% of users can be added per Tier 1 array. Scaling above this number,

additional arrays will need to be implemented. Additional capacity and performance capability is achieved

by adding larger disks or shelves, as appropriate, up to the controller’s performance limits. Each disk shelf


requires 1 hot spare per disk type. RAID is virtualized across all disks in an array (RAID10 or RAID6). Please

refer to the test methodology and results for specific workload characteristics. SSDs can be added for use

in scenarios where boot storms or provisioning speeds are an issue.

Controller Front-End IO Back-End IO Disk Shelf Disks SCOS (min)

2 x SC8000 (16GB)

2 x dual-port 8Gb FC cards (per controller)

2 x quad-port SAS cards (per controller)

2.5” SAS shelf (24 disks each)

2.5” 300GB 15K SAS (~206 IOPS each)

6.3

Tier 1 scaling guidance:

Users Controller Pairs

Disk Shelves 15K SAS Disks RAW Capacity Use

500 1 1 22 7TB T1 + T2

1000 1 2 48 15TB T1 + T2

2000 1 4 96 29TB T1 + T2

Figure 1 Example of a 1000 user Tier 1 array


3.3.3.2 Compellent Tier 2 Compellent Tier 2 storage is completely optional if a customer wishes to deploy discrete arrays for each

tier. The guidance below is provided for informational purposes and arrays built for this purpose will need

to be custom. The optional Compellent Tier 2 array consists of a standard dual controller configuration

and scales upward by adding disks and shelves. A single pair of SC8000 controllers should be able to

support Tier 2 for 10,000 basic users. Additional capacity and performance capability is achieved by adding

disks and shelves, as appropriate. Each disk shelf requires 1 hot spare per disk type. When designing for

Tier 2, capacity requirements will drive higher overall array performance capabilities due to the amount of

disk that will be on hand. Our base Tier 2 sizing guidance is based on 1 IOPS and 5GB per user.

Controller Front-End IO Back-End IO Disk Shelf Disks

2 x SC8000 (16GB) 2 x dual-port 8Gb FC cards (per controller)

2 x quad-port SAS cards (per controller)

2.5” SAS shelf (24 disks each)

2.5” 1TB NL SAS (~76 IOPS each)

Sample Tier 2 scaling guidance:

Users Controller Pairs Disk Shelves Disks RAW Capacity

500 1 1 7 7TB

1000 1 1 14 14TB

5000 1 3 66 66TB

10,000 1 6 132 132TB

Figure 2 Example of a 1000 user Tier 2 array.


3.3.4 NAS

3.3.4.1 FS7600

Model Features Scaling Uses

EqualLogic FS7600

Dual active-active controllers

24GB cache per controller (cache mirroring)

SMB & NFS support

AD-integration

Up to 2 FS7600 systems in a NAS cluster (4 controllers)

1Gb iSCSI via 16 x Ethernet ports

Each controller can support 1500 concurrent users

Up to 6000 total users in a 2 system NAS cluster

Scale out NAS for Local Tier 1 to provide file share HA

3.3.4.2 FS8600

Model Features Scaling Uses

Compellent FS8600

Dual active-active controllers

24GB cache per controller (cache mirroring)

SMB & NFS support

AD-integration

Up to 4 FS8600 systems in a NAS cluster (8 controllers)

Fibre Channel only

Each controller can support 1500 concurrent users

Up to 12,000 total users in a 4 system NAS cluster

Scale out NAS for Shared Tier 1 on Compellent, to provide file share HA (FC Only)

3.3.4.3 PowerVault NX3300 NAS Model Features Options Uses

PowerVault NX3300

Cluster-ready NAS built on Microsoft Windows Storage Server 2008 R2 Enterprise Edition

1 or 2 CPUs

1Gb and 10Gb NICs (configurable)

Scale out NAS for Shared Tier 1 on EqualLogic or Compellent, to provide file share HA (iSCSI).


3.4 Wyse Cloud Clients The following Wyse Cloud Clients are the recommended choices for this solution.

3.4.1 Wyse 5020-P25 Uncompromising computing with the benefits of secure, centralized

management. The Wyse 5020-P25 PCoIP zero client for VMware View is a secure,

easily managed zero client that provides outstanding graphics performance for

advanced applications such as CAD, 3D solids modeling, video editing and

advanced worker-level office productivity applications. Smaller than a typical notebook, this dedicated

zero client is designed specifically for VMware View. It features the latest processor technology from

Teradici to process the PCoIP protocol in silicon and includes client-side content caching to deliver the

highest level of performance available over 2 HD displays in an extremely compact, energy-efficient form

factor. The Wyse 5020-P25 delivers a rich user experience while resolving the challenges of provisioning,

managing, maintaining and securing enterprise desktops.

3.4.2 Wyse 5012-D10DP The Wyse 5012-D10DP is a high-performance and secure ThinOS 8 thin client that is

absolutely virus and malware immune. Combining the performance of a dual core AMD G-

Series APU with an integrated graphics engine and ThinOS, the 5012-D10DP offers

exceptional thin client PCoIP processing performance for VMware Horizon View

environments that handles demanding multimedia apps with ease and delivers brilliant

graphics. Powerful, compact and extremely energy efficient, the 5012-D10DP is a great VDI

end point for organizations that need high-end performance but face potential budget

limitations.

3.4.3 Wyse 7020-P45 Uncompromising computing with the benefits of secure, centralized management. The

Wyse 7020-P45 PCoIP zero client for VMware View is a secure, easily managed zero client

that provides outstanding graphics performance for advanced applications such as CAD, 3D

solids modeling, video editing and advanced worker-level office productivity applications.

About the size of a notebook, this dedicated zero client designed specifically for VMware

View. It features the latest processor technology from Teradici to process the PCoIP

protocol in silicon and includes client-side content caching to deliver the highest level of

display performance available over 4 HD displays in a compact, energy-efficient form

factor. The Wyse 7020-P45 delivers a rich user experience while resolving the challenges of provisioning,

managing, maintaining and securing enterprise desktops.


3.4.4 Wyse 7250-Z50D Designed for power users, the Wyse 7250-Z50D is the highest performing thin client on

the market. Highly secure and ultra-powerful, the 7250-Z50D combines Wyse-enhanced

SUSE Linux Enterprise with dual-core AMD 1.65 GHz processor and a revolutionary

unified engine for an unprecedented user experience. The 7250-Z50D eliminates

performance constraints for high-end, processing-intensive applications like computer-

aided design, multimedia, HD video and 3D modelling.

3.4.5 Wyse 7290-Z90D7 This is super high performance Windows Embedded Standard 7 thin client for virtual desktop

environments. Featuring a dual core AMD processor and a revolutionary unified engine that

eliminates performance constraints, the Wyse 7290-Z90D7 achieves incredible speed and

power for the most demanding embedded windows applications, rich graphics and HD video.

With touch screen capable displays, the Wyse 7290-Z90D7 adds the ease of an intuitive multi

touch user experience and is an ideal thin client for the most demanding virtual desktop

workload applications.

3.4.6 Wyse 7490-Z90Q8 The Wyse Z class is for users that demand more from their virtual desktop environments, yet

still need the security and management benefits of cloud clients. Featuring a quad core AMD

G-Series APUs, the Z class offers uncompromising performance with fast, flexible user

connectivity and outstanding energy-efficiency. The most demanding users in virtually any

VDI environment will appreciate the Z class power for challenging Windows® virtual

desktop and cloud applications, rich content creation and consumption, HD video, unified

communications and 3D graphics. The Z class is available as a thin client with Windows® 8

Embedded Standard operating systems.

3.4.7 Dell Chromebook 11 With its slim design and high performance, the Dell Chromebook 11 features a 4th

Generation Intel Celeron 2955U processor, 11.6-inch screen, up to 10-hours of

battery life and 16GB embedded Solid State Drive which allow it to book in

seconds. The Dell Chromebook 11 is available in two models with either 2GB or 4GB

of internal DDR3 RAM. This provides options for the education ecosystem, allowing

students, teachers and administrators to access, create and collaborate throughout

the day at a price point that makes widespread student computing initiatives affordable. The Dell

Chromebook 11 features an 11.6-inch, edge-to-edge glass screen that produces exceptional viewing

clarity at a maximum resolution of 1366x768 and is powered by Intel HD Graphics. The high-performing

display coupled with a front-facing 720p webcam creates exciting opportunities for collaborative learning.

The Dell Chromebook 11 is less than one inch in height and starts at 2.9lbs, making it highly portable. With


two USB 3.0 ports, Bluetooth 4.0 and an HDMI port, end users have endless possibilities for collaborating,

creating, consuming and displaying content. With battery life of up to 10-hours, the Chromebook is

capable of powering end users throughout the day.

Finally with a fully compliant HTML5 browser, the Dell Chromebook11 is an excellent choice as an

endpoint to a HTML5/BLAST connect Horizon View VDI desktop.


4 Software components

4.1 What's new in this release of Horizon View 6.0? This new release of VMware Horizon View delivers following important new features and enhancements:

RemoteApp – RemoteApp enables administrators to make programs that are accessed remotely through

a Remote Desktop Session (RDS) server appear as if they are running on the client computer versus a

remote desktop.

Virtual SAN – Horizon 6 with VMware Virtual SAN™ is a new storage technology that automates storage

provisioning and pools together with server-attached flash drives and hard disks and virtualizes them into

reliable storage. Built into the vSphere platform, the technology offers greater performance while

simplifying storage management. Virtual SAN eliminates the need to overprovision storage to ensure that

end users have enough IOPS per desktop.

Cloud pod architecture – The cloud pod architecture allows organizations to dynamically move and

locate View pods across multiple data centers for efficient management of end users across distributed

locations.

vDGA and vSGA 3D graphics enhancements – 3D graphics capabilities are enhanced to augment a

graphically rich user experience. Using Virtual Dedicated Graphics Acceleration (vDGA), a single virtual

machine is mapped to one physical graphics processing unit (GPU) in the ESXi host, providing high-end,

hardware-accelerated workstation graphics. Using Virtual Shared Graphics Acceleration (vSGA), multiple

virtual machines leverage physical GPUs that are installed locally in ESXi hosts, providing hardware

accelerated 3D graphics to multiple virtual desktops.

Unity Touch enhancements – Enhancements to VMware Unity Touch technology make it easier to

connect to View Connection Server or a View security server, log in to remote desktops in the data center,

and edit the list of connected servers. Unity Touch for VMware Horizon Client makes it easier to run

Windows apps on iPhone, iPad, and Android devices.

Additional OS support – View Connection Server, security server, and View Composer are supported on

Windows Server 2012 R2 operating systems.

Horizon View logs – Ability to send Horizon View logs to a Syslog server such as VMware vCenter Log

Insight.

Horizon View Agent – The Remote Experience Agent is now integrated with View Agent. Previously, you

had to install View Agent and the Remote Experience Agent to use features such as HTML Access, Unity

Touch, Real-Time Audio-Video, and Windows 7 Multimedia Redirection. In this release these features are

available by installing just the View Agent.


4.2 VMware Horizon View The solution is based on VMware Horizon View which provides a complete end-to-end solution delivering

Microsoft Windows virtual desktops to users on a wide variety of endpoint devices. Virtual desktops are

dynamically assembled on demand, providing users with pristine, yet personalized, desktops each time

they log on.

VMware Horizon View provides a complete virtual desktop delivery system by integrating several

distributed components with advanced configuration tools that simplify the creation and real-time

management of the virtual desktop infrastructure. For the complete set of details, please see the Horizon

View resources page at http://www.vmware.com/products/horizon-view/resources.html

The core Horizon View components include:

View Connection Server (VCS) – Installed on servers in the data center and brokers client connections,

The VCS authenticates users, entitles users by mapping them to desktops and/or pools, establishes secure

connections from clients to desktops, support single sign-on, sets and applies policies, acts as a DMZ

security server for outside corporate firewall connections and more.

View Client – Installed on endpoints. Is software for creating connections to View desktops that can be

run from tablets, Windows, Linux, or Mac PCs or laptops, thin clients and other devices.

View Portal – A web portal to access links for downloading full View clients. With HTML Access Feature

enabled enablement for running a View desktop inside a supported browser is enabled.

View Agent – Installed on all VMs, physical machines and Terminal Service servers that are used as a

source for View desktops. On VMs the agent is used to communicate with the View client to provide

services such as USB redirection, printer support and more.

View Administrator – A web portal that provides admin functions such as deploy and management of

View desktops and pools, set and control user authentication and more.

View Composer – This software service can be installed standalone or on the vCenter server and provides

enablement to deploy and create linked clone desktop pools (also called non-persistent desktops).

vCenter Server – This is a server that provides centralized management and configuration to entire virtual

desktop and host infrastructure. It facilitates configuration, provision, management services. It is installed

on a Windows Server 2008 host (can be a VM).

View Transfer Server – Manages data transfers between the data center and the View desktops that are

checked out on the end users’ desktops in offline mode. This Server is required to support desktops that

run the View client with Local Mode options. Replications and syncing are the functions it will perform

with offline images.

http://www.vmware.com/products/horizon-view/resources.html


4.3 VDI hypervisor platform

4.3.1 VMware vSphere 5 VMware vSphere 5 (currently vSphere 5.5 U2) is a virtualization platform used for building VDI and cloud

infrastructures. vSphere 5 represents a migration from the ESX architecture to the ESXi architecture.

VMware vSphere 5 includes three major layers: Virtualization, Management and Interface. The

Virtualization layer includes infrastructure and application services. The Management layer is central for

configuring, provisioning and managing virtualized environments. The Interface layer includes the vSphere

client and the vSphere web client.

Throughout the Dell Wyse Datacenter solution, all VMware best practices and prerequisites are adhered to

(NTP, DNS, Active Directory, etc.). The vCenter 5 VM used in the solution will be a single Windows Server

2012 R2 VM (Check for current Windows Server OS compatibility at:

http://www.vmware.com/resources/compatibility ), residing on a host in the management tier. SQL server

is a core component of vCenter and will be hosted on another VM also residing in the management tier.

All additional Horizon View components need to be installed in a distributed architecture, 1 role per VM.

For more information on VMware vSphere, visit http://www.vmware.com/products/vsphere

http://www.vmware.com/resources/compatibility

http://www.vmware.com/products/vsphere


5 Solution architecture

5.1 Compute server infrastructure

5.1.1 Local Tier 1 – Rack In the Local Tier 1 model, VDI sessions execute on local storage on each Compute server. Due to the local

disk requirement in the Compute layer, this model supports rack servers only. vSphere is used as the

solution hypervisor. In this model, only the Management server hosts access iSCSI storage to support the

solution’s Management role VMs. Because of this, the Compute and Management servers are configured

with different add-on NICs to support their pertinent network fabric connection requirements. Refer to

section 2.4.3.2 for cabling implications. The Management server host has reduced RAM and CPU and does

not require local disk space to host the management VMs.

Local Tier 1 Compute Host PowerEdge R730

Local Tier 1 Management Host PowerEdge R730

2 x Intel Xeon E5-2697v3 Processor (2.6Ghz) 2 x Intel Xeon E5-2660v3 Processor (2.6Ghz)

384GB Memory (24 x 16GB RDIMMs, 2133MT/s) 256GB Memory (16 x 16GB RDIMMs, 2133MT/s)

VMware vSphere on internal 8GB Dual SD VMware vSphere on internal 8GB Dual SD

10 x 300GB SAS 6Gbps 15k Disks (VDI)

PERC H730 Integrated RAID Controller – RAID10 Broadcom 57810 10Gb DP (iSCSI)

Broadcom 5720 1Gb QP NDC (LAN) Broadcom 57800 10Gb QP (LAN/iSCSI)

Broadcom 5720 1Gb DP NIC (LAN) Broadcom 5720 1Gb DP NIC (LAN)

iDRAC8 Enterprise iDRAC8 Enterprise

2 x 750W PSUs 2 x 750W PSUs

5.1.2 Shared Tier 1 – Rack In the Shared Tier 1 model, VDI sessions execute on shared storage so there is no need for local disk on

each server to host VMs. To provide server-level network redundancy using the fewest physical NICs

possible, both the Compute and Management servers use a split QP NDC: 2 x 10Gb ports for iSCSI, 2 x

1Gb ports for LAN. 2 additional DP NICs (2 x 1Gb + 2 x 10Gb) provide slot and connection redundancy for

both network fabrics. All configuration options are identical except for CPU and RAM which are reduced

on the Management host.


5.1.2.1 iSCSI

Shared Tier 1 Compute Host PowerEdge R730

Shared Tier 1 Management Host PowerEdge R730




Broadcom 57810 10Gb DP (iSCSI) Broadcom 57810 10Gb DP (iSCSI)

Broadcom 57800 10Gb QP (LAN/iSCSI) Broadcom 57800 10Gb QP (LAN/iSCSI)

Broadcom 5720 1Gb DP NIC (LAN) Broadcom 5720 1Gb DP NIC (LAN)



5.1.2.2 Fibre Channel

Shared Tier 1 Compute Host PowerEdge R730

Shared Tier 1 Management Host PowerEdge R730




1 x Broadcom 5720 1Gb QP NDC (LAN) 1 x Broadcom 5720 1Gb QP NDC (LAN)

1 x Broadcom 5720 1Gb DP NIC (LAN) 1 x Broadcom 5720 1Gb DP NIC (LAN)

2 x QLogic 2562 8Gb DP FC HBA 2 x QLogic 2562 8Gb DP FC HBA



In the above configurations, the R730-based Dell Wyse Datacenter Solution can support the following

user counts per server:

Local / Shared Tier 1 – Rack – User Densities

Workload Win 7 Win 8 Win 8.1

Standard 185* 140* 180

Enhanced 130* 112* 120

Professional 115* 90* 90

(*) Values based on R720 density testing. All others based on R730 density testing.

5.1.3 Shared Tier 1 – Blade The Dell M1000e Blade Chassis combined with the M620 blade server is the platform of choice for a high-

density data center configuration. The M620 is a feature-rich, dual-processor, half-height blade server

which offers a blend of density, performance, efficiency and scalability. The M620 offers remarkable

computational density, scaling up to 16 cores, 2 socket Intel Xeon processors and 24 DIMMs (768GB RAM)

of DDR3 memory in an extremely compact half-height blade form factor.


5.1.3.1 iSCSI

Shared Tier 1 Compute Host PowerEdge M620

Shared Tier 1 Management Host PowerEdge M620

2 x Intel Xeon E5-2690v2 Processor (3Ghz) 2 x Intel Xeon E5-2670v2 Processor (2.5Ghz)

256GB Memory (16 x 16GB DIMMs @ 1600Mhz) 96GB Memory (6 x 16GB DIMMs @ 1600Mhz)

VMware vSphere on 2 x 1GB internal SD VMware vSphere on 2 x 1GB internal SD

Broadcom 57810-k 10Gb DP KR NDC (iSCSI) Broadcom 57810-k 10Gb DP KR NDC (iSCSI)

1 x Intel i350 1Gb QP SERDES mezzanine (LAN) 1 x Intel i350 1Gb QP SERDES mezzanine (LAN)

iDRAC7 Enterprise w/ vFlash, 8GB SD iDRAC7 Enterprise w/ vFlash, 8GB SD

5.1.3.2 Fibre Channel Fibre Channel can be optionally leveraged as the block storage protocol for Compute and Management

hosts with Compellent Tier 1 and Tier 2 storage. Aside from the use of FC HBAs to replace the 10Gb NICs

used for iSCSI, the rest of the server configurations are the same. Please note that FC is only currently

supported using vSphere.

Shared Tier 1 Compute Host PowerEdge M620

Shared Tier 1 Management Host PowerEdge M620

2 x Intel Xeon E5-2690v2 Processor (3Ghz) 2 x Intel Xeon E5-2670v2 Processor (2.5Ghz)

256GB Memory (16 x 16GB DIMMs @ 1600Mhz) 96GB Memory (6 x 16GB DIMMs @ 1600Mhz)

VMware vSphere on 2 x 1GB internal SD VMware vSphere on 2 x 1GB internal SD

Broadcom 57810-k 10Gb DP KR NDC (LAN) Broadcom 57810-k 10Gb DP KR NDC (LAN)

1 x QLogic QME2572 8Gb FC mezzanine (FC) 1 x QLogic QME2572 8Gb FC mezzanine (FC)

iDRAC7 Enterprise w/ vFlash, 8GB SD iDRAC7 Enterprise w/ vFlash, 8GB SD

In the above configuration, the M620-based Dell Wyse Datacenter Solutions can support the following

single server user densities:

Shared Tier 1 – Blade – User Densities

Workload Win 7 Win 8 Win 8.1

Standard 185 140 150

Enhanced 130 112 105

Professional 115 90 93

Note: All values based on M620 density testing.

5.2 Management server infrastructure The Management role requirements for the base solution are summarized below. Use data disks for role-

specific application files and data, logs, IIS web files, etc. in the Management volume. Present Tier 2

volumes with a special purpose (called out above) in the format specified below:


Role vCPU RAM (GB) NIC OS + Data vDisk (GB)

Tier 2 Volume (GB)

VMware vCenter 2 8 1 40 + 5 100 (VMDK)

View Connection Server 2 8 1 40 – 5 -

SQL Server 5 8 1 40 + 5 210 (VMDK)

File Server 1 4 1 40 + 5 2048 (RDM)

Total 7 28 4 180 2358

5.2.1 SQL databases The VMware databases will be hosted by a single dedicated SQL 2012 SP1 Server VM (check DB

compatibility at: http://partnerweb.vmware.com/comp_guide2/sim/interop_matrix.php? ) in the

Management layer. Use caution during database setup to ensure that SQL data, logs and TempDB are

properly separated onto their respective volumes. Create all Databases that will be required for:

View Connection Server

vCenter

Initial placement of all databases into a single SQL instance is fine unless performance becomes an issue,

in which case database need to be separated into separate named instances. Enable auto-growth for each

DB.

Best practices defined by VMware are to be adhered to, to ensure optimal database performance.

The EqualLogic PS series arrays utilize a default RAID stripe size of 64K. To provide optimal performance,

configure disk partitions to begin from a sector boundary divisible by 64K.

Align all disks to be used by SQL Server with a 1024K offset and then formatted with a 64K file allocation

unit size (data, logs and TempDB).

5.2.2 DNS DNS plays a crucial role in the environment not only as the basis for Active Directory but will be used to

control access to the various VMware software components. All hosts, VMs and consumable software

components need to have a presence in DNS, preferably via a dynamic and AD-integrated namespace.

Microsoft best practices and organizational requirements are to be adhered to.

Pay consideration for eventual scaling, access to components that may live on one or more servers (SQL

databases, VMware services) during the initial deployment. Use CNAMEs and the round robin DNS

mechanism to provide a front-end “mask” to the back-end server actually hosting the service or data

source.

5.2.2.1 DNS for SQL To access the SQL data sources, either directly or via ODBC, a connection to the server name\ instance

name must be used. To simplify this process, as well as protect for future scaling (HA), instead of

connecting to server names directly, alias these connections in the form of DNS CNAMEs. So instead of

http://partnerweb.vmware.com/comp_guide2/sim/interop_matrix.php


connecting to SQLServer1\<instance name> for every device that needs access to SQL, the preferred

approach would be to connect to <CNAME>\<instance name>.

For example, the CNAME “VDISQL” is created to point to SQLServer1. If a failure scenario was to occur and

SQLServer2 would need to start serving data, we would simply change the CNAME in DNS to point to

SQLServer2. No infrastructure SQL client connections would need to be touched.

5.3 Scaling guidance Each component of the solution architecture scales independently according to the desired number of

supported users. Using the new Intel Ivy Bridge CPUs, rack and blade servers now scale equally from a

compute perspective.

The components can be scaled either horizontally (by adding additional physical and virtual

servers to the server pools) or vertically (by adding virtual resources to the infrastructure)

Eliminate bandwidth and performance bottlenecks as much as possible

Allow future horizontal and vertical scaling with the objective of reducing the future cost of

ownership of the infrastructure.

Component Metric Horizontal Scalability Vertical Scalability

Virtual Desktop Host/Compute Servers

VMs per physical host Additional hosts and clusters added as necessary

Additional RAM or CPU compute power

View Composer Desktops per instance Additional physical servers added to the Management cluster to deal with additional management VMs.

Additional RAM or CPU compute power

View Connection Servers

Desktops per instance Additional physical servers added to the Management cluster to deal with additional management VMs.

Additional VCS Management VMs.

VMware vCenter VMs per physical host and/or ESX hosts per vCenter instance

Deploy additional servers and use linked mode to optimize management

Additional vCenter Management VMs.

Database Services Concurrent connections, responsiveness of reads/ writes

Migrate databases to a dedicated SQL server and increase the number of management nodes

Additional RAM and CPU for the management nodes


File Services Concurrent connections, responsiveness of reads/ writes

Split user profiles and home directories between multiple file servers in the cluster. File services can also be migrated to the optional NAS device to provide high availability.

Additional RAM and CPU for the management nodes

The following tables indicate the server platform, desktop OS, hypervisor and delivery mechanism

5.3.1 Windows 7 – vSphere

Rack or Blade, Win7, vSphere

Standard User

Count

Enhanced User

Count

Professional User Count

Physical Mgmt. Servers

Physical Host

Servers

M1000e Blade

Chassis

View Conn.

Servers

Virtual vCenter Server

185 130 115 1 1 1 1 1

500 390 345 1 3 1 1 1

1000 780 690 2 6 1 1 1

2000 1430 1265 2 11 1 1 1

3000 2210 1955 2 17 2 2 1

4000 2860 2530 3 22 2 2 1

5000 3640 3220 3 28 2 3 1

6000 4290 3795 4 33 3 3 1

7000 4940 4370 4 38 3 4 1

8000 5720 5060 4 44 3 4 1

9000 6370 5635 4 49 4 5 1

10,000 7150 6325 4 55 4 5 1

Note: All values based on R720 and M620 density testing.


5.3.2 Windows 8 – vSphere

Rack or Blade, Win8, vSphere

Standard User

Count

Enhanced User

Count



Physical Host

Servers

M1000e Blade

Chassis

View Conn.

Servers


140 112 90 1 1 1 1 1

500 448 360 1 4 1 1 1

1000 896 720 2 8 1 1 1

2000 1680 1350 2 15 2 1 1

3000 2464 1980 2 22 2 2 1

4000 3248 2610 3 29 2 2 1

5000 4032 3240 3 36 3 3 1

6000 4816 3870 4 43 3 3 1

7000 5600 4500 4 50 4 4 1

8000 6496 5220 4 58 4 4 1

9000 7280 5850 4 65 5 5 1

10,000 8064 6480 4 72 5 5 1



5.3.3 Windows 8.1 – vSphere

Rack or Blade, Win8.1, vSphere

Standard User

Count

Enhanced User

Count



Physical Host

Servers

M1000e Blade

Chassis

View Conn.

Servers


180* 120* 90* 1 1 1 1 1

500 420 372 1 4 1 1 1

1000 735 651 2 7 1 1 1

2000 1470 1302 2 14 1 1 1

3000 2100 1860 2 20 2 2 1

4000 2835 2511 3 27 2 2 1

5000 3570 3162 3 34 3 3 1

6000 4200 3720 4 40 3 3 1

7000 4935 4371 4 47 4 4 1

8000 5670 5022 4 54 4 4 1

9000 6300 5580 4 60 4 5 1

10,000 7035 6231 4 67 5 5 1

(*) Values based on R730 density testing. All others based on R720 and M620 density testing.

5.3.4 Windows 2008R2 – vSphere

Rack or Blade, Win2008R2, vSphere

Standard User

Count

Enhanced User

Count



Physical Host

Servers

M1000e Blade

Chassis

View Conn.

Servers


213 150 132 1 1 1 1 1

500 450 396 1 3 1 1 1

1000 750 660 2 5 1 1 1

2000 1500 1320 2 10 1 1 1

3000 2250 1980 2 15 2 2 1

4000 2850 2508 3 19 2 2 1

5000 3600 3168 3 24 2 3 1

6000 4350 3828 4 29 3 3 1

7000 4950 4356 4 33 3 4 1

8000 5700 5016 4 38 3 4 1

9000 6450 5676 4 43 3 5 1

10000 7050 6204 4 47 4 5 1



5.4 Storage architecture overview The Dell Wyse Datacenter solution has a wide variety of tier 1 and tier 2 storage options to provide

maximum flexibility to suit any use case. Customers have the choice to leverage best-of-breed iSCSI

solutions from EqualLogic or Fibre Channel solutions from Dell Compellent while being assured the

storage tiers of the Dell Wyse Datacenter solution will consistently meet or outperform user needs and

expectations.

5.4.1 Local Tier 1 storage Choosing the local tier 1 storage option means that the virtualization host servers use ten (10) locally

installed hard drives to house the user desktop VMs. In this model, tier 1 storage exists as local hard disks

on the Compute hosts themselves. To achieve the required performance level, RAID 10 is recommended

for use across all local disks. A single volume per local tier 1 Compute host is sufficient to host the

provisioned desktop VMs along with their respective write caches.

5.4.2 Shared Tier 1 storage Choosing the shared tier 1 option means that the virtualization compute hosts are deployed in a diskless

mode and all leverage shared storage hosted on a high performance Dell storage array. In this model,

shared storage will be leveraged for tier 1 used for VDI execution and write cache. Based on the heavy

performance requirements of tier 1 VDI execution, it is recommended to use separate arrays for tier 1 and

tier 2 above 500 users for EqualLogic and above 1000 users for Compellent. It is recommended using

500GB LUNs for VDI and running 125 VMs per volume to minimize disk contention. Sizing to 1000 basic

users, for example, we will require 8 x 500GB volumes per array. A VMware Horizon View replica to

support a 1 to 500 desktop VM ratio should be located in a dedicated Replicas volume.

Volumes Size (GB) Storage Array Purpose File System

VDI-BaseImages 100 GB Tier 1 Storage for Base image for VDI

deployment VMFS

VDI-Replicas 100 GB Tier 1 Storage for Replica images created by

Horizon View VMFS

VDI-Images1 500 GB Tier 1 125 VMs-Storage for VDI Virtual

Machines in Horizon View Cluster VMFS
















For Shared Storage use on Compellent storage it is assumed that all pre-work for the setup of a properly

tiered architecture has been done to ensure proper data progression and optimal performance. General

guidance for configuration is as follows:

Replica (read only data) – SSD

User non-persistent - 15K

User Persistent - Data progression 15K --> 7K

Infrastructure volumes - Data progression "All tiers" (or) 15K --> 7K

5.4.3 Shared Tier 2 storage Tier 2 is shared iSCSI or FC storage used to host the Management server VMs and user data. EqualLogic

PS4100E series 1Gb arrays can be used for smaller scale deployments (Local Tier 1 only), the PS62x0XS or

PS65x0XS series for larger deployments (up to 16 in a group), or a single CML array scaled up to 10K users.

The 10Gb iSCSI variants are intended for use in Shared Tier 1 solutions. The Compellent Tier 2 array, as

specified in section 3.3.2 scales simply by adding disks. The table below outlines the volume requirements

for Tier 2. Larger disk sizes can be chosen to meet the capacity needs of the customer. The user data can

be presented either via a file server VM using RDM for small scale deployments or via NAS for large scale or

HA deployments. The solution as designed presents all SQL disks using VMDK formats. RAID 50 can be

used in smaller deployments but is not recommended for critical environments. The recommendation for

larger scale and mission critical deployments with higher performance requirements is to use RAID 10 or

RAID 6 to maximize performance and recoverability. The following depicts the component volumes

required to support a 500 user environment. Additional Management volumes can be created as needed

along with size adjustments as applicable for user data and profiles.

Volumes Size (GB) Storage Array Purpose File System

Management 350 Tier 2 vCenter, View Connection Server, File and SQL

VMFS

User Data 2048 Tier 2 File Server/ NAS RDM/NTFS

User Profiles 20 Tier 2 User profiles VMFS

SQL DATA 100 Tier 2 SQL VMFS

SQL LOGS 100 Tier 2 SQL VMFS

TempDB Data 5 Tier 2 SQL VMFS

TempDB Logs 5 Tier 2 SQL VMFS

SQL Witness 1 Tier 2 SQL (optional) VMFS

Templates/ ISO 200 Tier 2 ISO storage (optional) VMFS

5.4.4 Storage networking – EqualLogic iSCSI Dell’s iSCSI technology provides compelling price/performance in a simplified architecture while

improving manageability in virtualized environments. Specifically, iSCSI offers virtualized environments


simplified deployment, comprehensive storage management and data protection functionality and

seamless VM mobility. Dell iSCSI solutions give customers the “Storage Direct” advantage – the ability to

seamlessly integrate virtualization into an overall, optimized storage environment.

If iSCSI is the selected block storage protocol, then the Dell EqualLogic MPIO plugin is installed on all

hosts that connect to iSCSI storage. This module is added via a command line using a Virtual Management

Appliance (vMA) from VMware. This plugin allows for easy configuration of iSCSI on each host. The MPIO

plugin allows the creation of new or access to existing data stores and handle IO load balancing. The

plugin will also configure the optimal multi-path settings for the data stores as well. Some key settings to

be used as part of the configuration:

Specify 2 IP Addresses for iSCSI on each host

Specify NICs

Specify Jumbo Frames at 9000 MTU

Initialize iSCSI initiator

Specify IP for the EqualLogic Storage group


5.4.5 Storage networking – Compellent Fibre Channel Based on Fluid Data architecture, the Dell Compellent Storage Center SAN provides built-in intelligence

and automation to dynamically manage enterprise data throughout its lifecycle. Together, block-level

intelligence, storage virtualization, integrated software and

modular, platform-independent hardware enable exceptional

efficiency, simplicity and security.

Storage Center actively manages data at a block level using real-

time intelligence, providing fully virtualized storage at the disk

level. Resources are pooled across the entire storage array. All

virtual volumes are thin-provisioned and with sub-LUN tiers,

data is automatically moved between tiers and RAID levels based

on actual use.

If Fibre Channel is the selected block storage protocol, then the

Compellent Storage Center Integrations for VMware vSphere

client plug-in is installed on all hosts. This plugin enables all

newly created data stores to be automatically aligned at the

recommended 4MB offset. Although a single fabric can be configured to begin with to reduce costs, as a

best practice recommendation, the environment needs to be configured with 2 fabrics to provide multi-

path and end-to-end redundancy.

Using QLogic HBAs the following BIOS settings were used:

Set the “connection options” field to 1 for point to point only

Set the “login retry count” field to 60 attempts

Set the “port down retry” count field to 60 attempts

Set the “link down timeout” field to 30 seconds

Set the “queue depth” (or “Execution Throttle”) field to 255

This queue depth can be set to 255 because the ESXi VMkernel driver module and DSNRO can

more conveniently control the queue depth

5.4.5.1 FC Zoning Zone at least 1 port from each server HBA to communicate with a single Compellent fault domain. The

result of this will be 2 distinct FC fabrics and 4 redundant paths per server. Round Robin or Fixed Paths are

supported. Leverage Compellent Virtual Ports to minimize port consumption as well as simplify

deployment. Zone each controller’s front-end virtual ports, within a fault domain, with at least one ESXi

initiator per server.

A Fabric

Compute & Mgmt hosts

FC switchA

B Fabric

SAN

FC switchB

HBAA

HBAB


5.5 Virtual networking

5.5.1 Local Tier 1 – Rack – iSCSI The network configuration in this model will vary between the Compute and Management hosts. The

Compute hosts will not need access to iSCSI storage since they are hosting VDI sessions locally. Since the

Management VMs will be hosted on shared storage, they can take advantage of VMware vMotion. The

following outlines the VLAN requirements for the Compute and Management hosts in this solution model:

Compute hosts (Local Tier 1)

o Management VLAN: Configured for hypervisor infrastructure traffic – L3 routed via core

switch

o VDI VLAN: Configured for VDI session traffic – L3 routed via core switch

Management hosts (Local Tier 1)

o Management VLAN: Configured for hypervisor Management traffic – L3 routed via core

switch

o vMotion VLAN: Configured for vMotion traffic – L2 switched only, trunked from Core

o iSCSI VLAN: Configured for iSCSI traffic – L2 switched only via ToR switch

o VDI Management VLAN: Configured for VDI infrastructure traffic – L3 routed via core

switch

An optional iDRAC VLAN can be configured for all hardware management traffic – L3 routed via

core switch

Following best practices, LAN and block storage traffic will be separated in solutions >1000 users. This

traffic can be combined within a single switch in smaller stacks to minimize buy-in costs. Each Local Tier 1

Compute host will have a quad port NDC as well as a 1Gb dual port NIC. Configure the LAN traffic from

the server to the ToR switch as a LAG.


5.5.1.1 vSphere

The Compute host will require 2 vSwitches, one for VDI LAN traffic and another for the ESXi Management.

Configure both vSwitches so that each is physically connected to both the onboard NIC as well as the

add-on NIC. Set all NICs and switch ports to auto negotiate.

The Management hosts have a slightly different configuration since they will additionally access iSCSI

storage. The add-on NIC for the Management hosts will be a 1Gb quad port NIC. 3 ports of both the NDC

and add-on NIC will be used for the required connections. Isolate iSCSI onto its own vSwitch with

redundant ports and connections from all 3 vSwitches. Connections should pass through both the NDC

and add-on NIC per the diagram below. Configure the LAN traffic from the server to the ToR switch as a

LAG.

vm nic0 1000 Fu ll

vm nic4 1000 Fu ll

Physica l A dapters

M gm t

VM kernel Port

Standard Sw itch: vSw itch0

V m k0: 10 .20 .1.50 | V LA N ID : 10

vm nic1 1000 Fu ll

vm nic5 1000 Fu ll

Physica l A dapters

VD I VLAN

Virtual M achine Port G roup


X v irtua l m achine(s) | V LA N ID : 5

C om pute | Loca l T ier 1

VD I-1

VD I-2

VD I-3

vsw 1

LAN

vsw 0

Mgmt

1 Gb QP NDC

R 730

1 Gb DP NIC

Compute Hosts – Local Tier 1

F 10 - LAN

ToR


vSwitch0 carries traffic for both Management and vMotion which needs be VLAN-tagged so that either

NIC can serve traffic for either VLAN. The Management VLAN will be L3 routable while the vMotion VLAN

will be L2 non-routable.

5.5.2 Shared Tier 1 – Rack – iSCSI The network configuration in this model is identical between the Compute and Management hosts. Both

need access to iSCSI storage since they are hosting VDI sessions from shared storage and both can

leverage vMotion as a result as well. The following outlines the VLAN requirements for the Compute and

Management hosts in this solution model:

Compute hosts (Shared Tier 1)


switch

iSCSI0

VM kernel Port

vm k2: 10 .1.1 .10 | V LA N ID : 11

vm nic2 1000 Fu ll

vm nic6 1000 Fu ll

Physica l A dapters

VD I M gm t VLAN




M gm t | Loca l T ier 1

iSCSI1

VM kernel Port

vm k3: 10 .1.1 .11 | V LA N ID : 11

SQ L

vCenter

File

vm nic0 10000 Fu ll

vm nic4 10000 Fu ll

Physica l A dapters


vm nic1 1000 Fu ll

vm nic5 1000 Fu ll

Physica l A dapters

M gm t

VM kernel Port


vm k0: 10 .20 .1.51 | V LA N ID : 10

VM otion

VM kernel Port

vm k1: 10 .1.1 .1 | V LA N ID : 12

vsw 2

LAN

vsw 1

iSCSI

R 730

Mgmt Hosts – Local Tier 1

vsw 0 Mgmt /

Migration

F 10 - LAN

F 10 - iSCSI 1 Gb QP NIC

1 Gb QP NDC

ToR





Management hosts (Shared Tier 1)


switch



o VDI Management VLAN: Configured for VDI infrastructure traffic – L3 routed via core

switch

An optional iDRAC VLAN can be configured for all hardware management traffic – L3 routed via

core switch

Following best practices, iSCSI and LAN traffic will be physically separated into discrete fabrics. Each

Shared Tier 1 Compute and Management host will have a quad port NDC (2 x 1Gb + 2 x 10Gb SFP+), a

10Gb dual port NIC, as well as a 1Gb dual port NIC. Isolate iSCSI onto its own vSwitch with redundant

ports. Connections from all 3 vSwitches should pass through both the NDC and add-on NICs per the

diagram below. Configure the LAN traffic from the server to the ToR switch as a LAG.

5.5.2.1 vSphere

vSwitch0 carries traffic for both Management and vMotion which needs to be VLAN-tagged so that either



F 10 - LAN

F 10 - iSCSI vsw 1

iSCSI

10 Gb

DP NIC

vsw 0 Mgmt /

Migration

vsw 2

LAN

R 730

Compute + Mgmt Hosts – Shared Tier 1 – iSCSI

1 Gb DP NIC

2 x 1 Gb

2 x 10 Gb QP NDC

ToR


The Management server is configured identically except for the VDI Management VLAN which is fully

routed but should be separated from the VDI VLAN used on the Compute host. Care should be taken to

ensure that all vSwitches are assigned redundant NICs that are NOT from the same PCIe device.

iSCSI0

VM kernel Port

vm k2: 10 .1.1 .10 | V LA N ID : 11

vm nic2 1000 Fu ll

vm nic6 1000 Fu ll

Physica l A dapters

VD I VLAN




C om pute | Shared T ier 1

iSCSI1

VM kernel Port

vm k3: 10 .1.1 .11 | V LA N ID : 11

VD I-1

VD I-2

VD I-3

vm nic0 10000 Fu ll

vm nic4 10000 Fu ll

Physica l A dapters


vm nic1 1000 Fu ll

vm nic5 1000 Fu ll

Physica l A dapters

M gm t

VM kernel Port


vm k0: 10 .20 .1.51 | V LA N ID : 10

VM otion

VM kernel Port

vm k1: 10 .1.1 .1 | V LA N ID : 12

iSCSI0

VM kernel Port

vm k2: 10 .1.1 .10 | V LA N ID : 11

vm nic2 1000 Fu ll

vm nic6 1000 Fu ll

Physica l A dapters

VD I M gm t VLAN




M gm t | Shared T ier 1

iSCSI1

VM kernel Port

vm k3: 10 .1.1 .11 | V LA N ID : 11

SQ L

vCenter

File

vm nic0 10000 Fu ll

vm nic4 10000 Fu ll

Physica l A dapters


vm nic1 1000 Fu ll

vm nic5 1000 Fu ll

Physica l A dapters

M gm t

VM kernel Port


vm k0: 10 .20 .1.51 | V LA N ID : 10

VM otion

VM kernel Port

vm k1: 10 .1.1 .1 | V LA N ID : 12


5.5.3 Shared Tier 1 – Rack – Fibre Channel Using Fibre Channel based storage eliminates the need to build iSCSI into the network stack but requires

additional fabrics to be built out. The network configuration in this model is identical between the

Compute and Management hosts. Both need access to FC storage since they are hosting VDI sessions

from shared storage and both can leverage vMotion as a result as well. The following outlines the VLAN

requirements for the Compute and Management hosts in this solution model:


o Management VLAN: Configured for hypervisor Management traffic – L3 routed via core switch






o VDI Management VLAN: Configured for VDI infrastructure traffic – L3 routed via core switch

An optional iDRAC VLAN can be configured for all hardware management traffic – L3 routed via core switch

FC and LAN traffic are physically separated into discrete switching fabrics. Each Shared Tier 1 Compute

and Management host will have a quad port NDC (4 x 1Gb), a 1Gb dual port NIC, as well as 2 x 8Gb dual

port FC HBAs. Connections from both vSwitches should pass through both the NDC and add-on NICs per

the diagram below. Configure the LAN traffic from the server to the ToR switch as a LAG.

5.5.3.1 vSphere




Compute + Mgmt Hosts – Shared Tier 1 – FC

F 10 - LAN

ToR

vsw 1

LAN

1 Gb

QP NDC

R 730

1 Gb

DP NIC

8 Gb

FC HBA

8 Gb

FC HBA

Brocade - FC

vsw 0 Mgmt /

migration



routed but should be separated from the VDI VLAN used on the Compute host.

5.5.4 Shared Tier 1 – Blade – iSCSI The network configuration in this model is identical between the Compute and Management hosts. The

following outlines the VLAN requirements for the Compute and Management hosts in this solution model:


vmnic4 1000 Full

vmnic5 1000 Full

Physical Adapters

VDI VLAN

Virtual Machine Port Group

Standard Switch: vSwitch2

X virtual machine(s) | VLAN ID: 6

Compute | Shared Tier 1 (FC)

VDI-1

VDI-2

VDI-3

vmnic2 1000 Full

vmnic3 1000 Full

Physical Adapters

Mgmt

VMkernel Port


vmk0: 10.20.1.51 | VLAN ID: 10

VMotionVMkernel Port

vmk1: 10.1.1.1 | VLAN ID: 12

vmnic4 1000 Full

vmnic5 1000 Full

Physical Adapters

VDI Mgmt VLAN

Virtual Machine Port Group


X virtual machine(s) | VLAN ID: 6

Mgmt | Shared Tier 1 (FC)

SQL

vCenter

File

vmnic2 1000 Full

vmnic3 1000 Full

Physical Adapters

Mgmt

VMkernel Port


vmk0: 10.20.1.51 | VLAN ID: 10

VMotionVMkernel Port

vmk1: 10.1.1.1 | VLAN ID: 12











An optional iDRAC VLAN can be configured for all hardware management traffic – L3 routed via core switch

Following best practices, iSCSI and LAN traffic will be physically separated into discrete fabrics. Each

Shared Tier 1 Compute and Management blade host will have a 10Gb dual port LOM in the A fabric and a

1Gb quad port NIC in the B fabric. 10Gb iSCSI traffic will flow through A fabric using 2 x IOA blade

interconnects. 1Gb LAN traffic will flow through the B fabric using 2 x M6348 blade interconnects. The C

fabric will be left open for future expansion. Connections from 10Gb and 1Gb traffic vSwitches should pass

through the blade mezzanines and interconnects per the diagram below. Configure the LAN traffic from

the server to the ToR switch as a LAG if possible.

5.5.4.1 vSphere






routed but should be separated from the VDI VLAN used on the Compute host.

5.5.5 Shared Tier 1 – Blade – Fibre Channel Using Fibre Channel based storage eliminates the need to build iSCSI into the network stack but requires

additional fabrics to be built out. The network configuration in this model is identical between the

iSCSI0

VM kernel Port

vm k2: 10 .1.1 .10 | V LA N ID : 11

vm nic4 1000 Fu ll

vm nic5 1000 Fu ll

Physica l A dapters

VD I M gm t VLAN




M gm t | Shared T ier 1

iSCSI1

VM kernel Port

vm k3: 10 .1.1 .11 | V LA N ID : 11

SQ L

vC enter

File

vm nic0 10000 Fu ll

vm nic1 10000 Fu ll

Physica l A dapters


vm nic2 1000 Fu ll

vm nic3 1000 Fu ll

Physica l A dapters

M gm t

VM kernel Port


vm k0: 10 .20 .1.51 | V LA N ID : 10

VM otion

VM kernel Port

vm k1: 10 .1.1 .1 | V LA N ID : 12


Compute and Management hosts. The following outlines the VLAN requirements for the Compute and

Management hosts in this solution model:









An optional DRAC VLAN can be configured for all hardware management traffic – L3 routed via core switch

FC and LAN traffic are physically separated into discrete switching fabrics. Each Shared Tier 1 Compute

and Management blade will have a 10Gb dual port LOM in the A fabric and an 8Gb dual port HBA in the B

fabric. All LAN and management traffic will flow through the A fabric using 2 x IOA blade interconnects

partitioned to the connecting blades. 8Gb FC traffic will flow through the B fabric using 2 x M5424 blade

interconnects. The C fabric will be left open for future expansion. Connections from the vSwitches and

storage fabrics should pass through the blade mezzanines and interconnects per the diagram below.

Configure the LAN traffic from the server to the ToR switch as a LAG.

5.5.5.1 vSphere

5.5.5.2 Shared Tier 1 – Blade – Network partitioning

Network partitioning (NPAR) takes place within the UEFI of the 10GB LOMs of each blade in the A fabric. Partitioning allows a 10Gb NIC to be split into multiple logical NICs that can be assigned differing amounts of bandwidth. 4 partitions are defined per NIC with the amounts specified below. We only require 2 partitions per NIC port so the unused partitions receive a bandwidth of 1. Partitions can be oversubscribed, but not the reverse. We will be partitioning out a total of 4 x 5Gb NICs with the remaining 4 unused. Use care to ensure that each vSwitch receives a NIC from each physical port for redundancy.


5.6 Solution high availability High availability (HA) is offered to protect each layers of the solution architecture, individually if desired.

Following the N+1 model, additional ToR switches for LAN, iSCSI, or FC are added to the Network layer

and stacked to provide redundancy as required, additional compute and management hosts are added to

their respective layers, vSphere clustering is introduced in the management layer, SQL is mirrored or

clustered, an F5 device can be leveraged for load balancing and a NAS device can be used to host file

shares. Storage protocol switch stacks and NAS selection will vary based on chosen solution architecture.

The HA options provides redundancy for all critical components in the stack while improving the

performance and efficiency of the solution as a whole.


An additional switch is added at the network tier which will be configured with the original as a

stack and equally spreading each host’s network connections across both.

At the compute tier an additional ESXi host is added to provide N+1 protection provided by

vSphere for computer tier protection. In a rack based solution with local tier 1 storage, there will

be no vSphere HA cluster in the compute tier as VMs that run here run on local disks.

A number of enhancements occur at the Management tier, the first of which is the addition of

another host. The Management hosts will then be configured in an HA cluster. All applicable

Horizon View server roles can then be duplicated on the new host where connections to each will

be load balanced via the addition of a F5 Load Balancer. SQL will also receive greater protection

through the addition and configuration of a SQL mirror with a witness.

5.6.1 Compute layer HA (Local Tier 1) The optional HA bundle adds an additional host in the Compute and Management layers to provide

redundancy and additional processing power to spread out the load. The Compute layer in this model

does not leverage shared storage so hypervisor HA does not provide a benefit here. If a single host fails,

another will need to be spun up in the cluster or extra server capacity can be pre-configured and running

in active status to handle the reconnection/startup of new desktops to accommodate the users from failed

host.

Because only the Management hosts have access to shared storage, in this model, only these hosts need

to leverage the full benefits of hypervisor HA. The Management hosts can be configured in an HA cluster

with or without the HA bundle. An extra server in the Management layer will provide protection should a

host fail.

vSphere HA Admission control can be configured one of three ways to protect the cluster. This will vary

largely by customer preference but the most manageable and predictable options are percentage

reservations or a specified hot standby. Reserving by percentage will reduce the overall per host density

capabilities but will make some use of all hardware in the cluster. Additions and subtractions of hosts will

require the cluster to be manually rebalanced. Specifying a failover host, on the other hand, will ensure

maximum per host density numbers but will result in hardware sitting idle.


5.6.2 vSphere HA (Shared Tier 1) Both compute and management hosts are identically configured, within their respective tiers and leverage

shared storage so can make full use of vSphere HA. The Compute hosts can be configured in an HA

cluster following the boundaries of vCenter with respect to limits imposed by VMware (3000 VMs per

vCenter). This will result in multiple HA clusters managed by multiple vCenter servers.

A single HA cluster will be sufficient to support the Management layer up to 10K users. An additional host

can be used as a hot standby or to thin the load across all hosts in the cluster.

5.6.3 Horizon View infrastructure protection VMware Horizon View infrastructure data protection with Dell Data Protection – http://dell.to/1ed2dQf

Compute Host Cluster

Manage 10000 VMs

vCenter

http://dell.to/1ed2dQf


5.6.4 Management server high availability The applicable core Horizon View roles will be load balanced via DNS by default. In environments

requiring HA, F5 can be introduced to manage load-balancing efforts. Horizon View, VCS and vCenter

configurations (optionally vCenter Update Manager) are stored in SQL which will be protected via the SQL

mirror.

If the customer desires, some Role VMs can be optionally protected further via the form of a cold stand-by

VM residing on an opposing management host. A vSphere scheduled task can be used, for example, to

clone the VM to keep the stand-by VM current. Note – In the HA option, there is no file server VM, its

duties have been replaced by introducing a NAS head.

The following will protect each of the critical infrastructure components in the solution:

The Management hosts will be configured in a vSphere cluster.

SQL Server mirroring is configured with a witness to further protect SQL.

5.6.5 Horizon View VCS high availability The VCS role as a VM and running in a VMware HA Cluster, the VCS server can be guarded against a

physical server failure.

For further protection in an HA configuration, deploy multiple replicated View Connection Server

instances in a group to support load balancing and HA. Replicated instances must exist on within a LAN

connection environment it is not recommended VMware best practice to create a group across a WAN or

similar connection.

5.6.6 Windows File Services high availability High availability for file services will be provided by the Dell FS7600, FS8600 or PowerVault NX3300

clustered NAS devices. To ensure proper redundancy, distribute the NAS cabling between ToR switches.

Unlike the FS8600, the FS7600 and NX3300 do not support for 802.1q (VLAN tagging) so configure the

connecting switch ports with native VLANs, both iSCSI and LAN/ VDI traffic ports. Best practice dictates

that all ports be connected on both controller nodes. The back-end ports are used for iSCSI traffic to the

storage array as well as internal NAS functionality (cache mirroring and cluster heart beat). Front-end ports

can be configured using Adaptive Load Balancing or a LAG (LACP).

The Dell Wyse Solutions Engineering recommendation is to configure the original file server VM to use

RDMs to access the storage LUNs, therefore migration to the NAS will be simplified by changing the

presentation of these LUNs from the file server VM to the NAS.


5.6.7 SQL Server high availability HA for SQL will be provided via a 3-server synchronous mirror

configuration that includes a witness (High safety with automatic

failover). This configuration will protect all critical data stored within the

database from physical server as well as virtual server problems. DNS will

be used to control access to the active SQL server, please refer to

section 5.7.1 for more details. Place the principal VM that will host the

primary copy of the data on the first Management host. Place the mirror

and witness VMs on the second or later Management hosts. Mirror all

critical databases to provide HA protection.

The following article details the step-by-step mirror configuration: http://www.sqlserver-

training.com/how-to-setup-mirroring-in-sql-server-screen-shots

Additional resources can be found in TechNet: http://technet.microsoft.com/en-

us/library/ms189047.aspx http://technet.microsoft.com/en-us/library/ms188712.aspx

5.6.8 Load balancing Depending on which management components are to be made highly available, the use of a load

balancer may be required. The following management components can use a load balancer to function in

a high availability mode:

View Connection Servers

VMware Security Servers (WAN connected VCS’)

Virtual Desktops (PCoIP traffic)

Dell recommends the F5 for load balancing the Dell Wyse Datacenter for VMware Horizon View solution.

For additional reference, please review the following document. In particular, page 44 has a good

overview and architecture example. http://www.f5.com/pdf/deployment-guides/vmware-view5-iapp-

dg.pdf.

5.6.8.1 DNS for load balancing When considering DNS for non SQL-based components such as VCS or file servers, where a load

balancing behavior is desired, invoke the native DNS round robin feature. To invoke round robin, enter the

resource records for a service into DNS as A records with the same name.

For example, in the base configuration the single VCS server will have its own hostname registered in DNS

as an A record. Create a new A record to be used should additional VCS’s come online or be retired for

whatever reason. This creates machine portability at the DNS layer to remove the importance of actual

server hostnames. The name of this new A record is unimportant but must be used as the primary name

record to gain access to the resource, not the server’s host name! In this case three new created A records

called “WebInterface”, all presumably pointing to three different servers.

http://www.sqlserver-training.com/how-to-setup-mirroring-in-sql-server-screen-shots

http://www.sqlserver-training.com/how-to-setup-mirroring-in-sql-server-screen-shots

http://technet.microsoft.com/en-us/library/ms189047.aspx



http://www.f5.com/pdf/deployment-guides/vmware-view5-iapp-dg.pdf

http://www.f5.com/pdf/deployment-guides/vmware-view5-iapp-dg.pdf


When a client requests the name Web Interface, DNS will direct them to the 3 hosts in round robin

fashion. The following resolutions were performed from 2 different clients. Repeat this method of creating

an identical but load-balanced namespace for all applicable components of the architecture stack.

5.7 VMware Horizon View communication flow


6 Customer-provided solution components

6.1 Customer-provided storage requirements In the event that a customer wishes to provide their own storage array solution for a Dell Wyse Datacenter

solution, the following minimum hardware requirements must be met:

Feature Minimum Requirement Notes

Total Tier 2 Storage Space User count and workload

dependent 1Gb/ 10Gb iSCSI or FC storage required on NL SAS disks minimally.

Tier 1 IOPS Requirement (Total Users) x workload IOPS

6-30 IOPS per user may be required depending on workload. T1 storage should be capable of providing user IOPS requirement concurrently to all hosted users.

Tier 2 IOPS Requirement (Total Users) x 1 IOPS File share usage and size of deployment may shift this requirement.

Data Networking

1GbE Ethernet for LAN/T2 iSCSI

10GbE Ethernet for T1 iSCSI

8Gb FC

Data networking traffic should be isolated on dedicated NICs and HBAs in each applicable host.

10GbE Ethernet for T1 iSCSI 10, 6

RAID 10 is leveraged for T1 local storage and can be used if required for shared T2. RAID 6 is used in shared T1 and can be optionally used for T2 as well.

6.2 Customer-provided switching requirements Feature Minimum Requirement Notes

Switching Capacity Line rate switch

1Gb or 10Gb switching pertinent to solution being implements. 1Gb switching for iSCSI is only suitable for T2. T1 iSCSI requires 10Gb.

10Gbps Ports Uplink to Core 10Gbps Ports

1Gbps Ports 5x per Management server

5x per Compute Server 6x per Storage Array

1Gbps Ports

VLAN Support IEEE 802.1Q tagging and

port-based VLAN support.

Stacking Capability Yes

The ability to stack switches into a consolidated management framework is preferred to minimize disruption and planning when up linking to core networks.


7 Solution performance and testing

7.1 Load generation and monitoring

7.1.1 VMware View Planner View Planner, currently at version 3, operates in two modes; Benchmark mode and Flexible mode.

Benchmark mode is used in a locked down or fixed mode and is designed to run a workload that cannot

be changed for the purpose of determining a standardized result on a given set of hardware. Flexible

mode is more of a traditional version designed to allow customization of workloads to try and emulate

target scenarios and produce results designed to determine the scale and density of a given infrastructure.

Similar to Login VSI, “helper” or “launcher” systems aid in the testing process with View Planner 3.0. This

tool is required for testing Dell Wyse Datacenter for VMware Horizon View Solutions in VMware’s Rapid

Desktop Program.

7.1.2 Login VSI – Login Consultants Login VSI is the de-facto industry standard tool for testing VDI environments and server-based computing

/ terminal services environments. It installs a standard collection of desktop application software (e.g.

Microsoft Office, Adobe Acrobat Reader etc.) on each VDI desktop; it then uses launcher systems to

connect a specified number of users to available desktops within the environment. Once the user is

connected the workload is started via a logon script which starts the test script once the user environment

is configured by the login script. Each launcher system can launch connections to a number of 'target'

machines (i.e. VDI desktops), with the launchers being managed by a centralized management console,

which is used to configure and manage the Login VSI environment. Important to note is that there are

some performance changes between VSI 3.7 and 4.0. Namely, desktop read/write IO has decreased a bit

while CPU utilization has increased in 4.0.

7.1.3 Liquidware Labs Stratusphere UX Stratusphere UX was used during each test run to gather data relating to User Experience and desktop

performance. Data was gathered at the Host and Virtual Machine layers and reported back to a central

server (Stratusphere Hub). The hub was then used to create a series of “Comma Separated Values” (.csv)

reports which have then been used to generate graphs and summary tables of key information. In addition

the Stratusphere Hub generates a magic quadrant style scatter plot showing the Machine and IO

experience of the sessions. The Stratusphere hub was deployed onto the core network therefore its


monitoring did not impact the servers being tested. This core network represents an existing customer

environment and also includes the following services:

Active Directory

DNS

DHCP

Anti-Virus

Stratusphere UX calculates the User Experience by monitoring key metrics within the Virtual Desktop

environment, the metrics and their thresholds are shown in the following screen shot:

7.1.4 EqualLogic SAN HQ EqualLogic SAN HQ was used for monitoring the Dell EqualLogic storage units in each bundle. SAN HQ

has been used to provide IOPS data at the SAN level; this has allowed the team to understand the IOPS

required by each layer of the solution. This report details the following IOPS information:

File Server IOPS for User Profiles and Home Directories

SQL Server IOPS required to run the solution databases

Infrastructure VM IOPS (the IOPS required to run all the infrastructure Virtual servers)

7.1.5 VMware vCenter VMware vCenter has been used for VMware vSphere-based solutions to gather key data (CPU, Memory

and Network usage) from each of the desktop hosts during each test run. This data was exported to .csv

files for each host and then consolidated to show data from all hosts. While the report does not include

specific performance metrics for the Management host servers, these servers were monitored during

testing and were seen to be performing at an expected performance level.


7.2 Performance analysis methodology In order to ensure the optimal combination of end user experience (EUE) and cost-per-user, performance

analysis and characterization (PAAC) on Dell Wyse Datacenter solutions is carried out using a carefully

designed, holistic methodology that monitors both hardware resource utilization parameters and EUE

during load-testing. This methodology is based on the three pillars shown below. Login VSI is currently the

load-testing tool used during PAAC of Dell Wyse Datacenter solutions; Login VSI is the de-facto industry

standard for VDI and server-based computing (SBC) environments and is discussed in more detail below.

7.2.1 Resource utilization Poor end user experience is one of the main risk factors when implementing desktop virtualization but the

root cause for poor end user experience is resource contention – hardware resources at some point in the

solution have been exhausted causing poor performance. In order to ensure that this has not happened

(and that it is not close to happening), PAAC on Dell Wyse Datacenter solutions monitors the relevant

resource utilization parameters and applies relatively conservative thresholds as shown in the table below.

As discussed above, these thresholds are carefully selected to deliver an optimal combination of good end

user experience and cost-per-user, while also providing burst capacity for seasonal / intermittent spikes in

usage. These thresholds are used to decide the number of virtual desktops (density) that can be hosted by

a specific hardware environment (i.e. combination of server, storage and networking) that forms the basis

for this Dell Wyse Datacenter for VMware Horizon View Reference Architecture.


Resource Utilization Thresholds

Parameter Pass / Fail Threshold

Physical host CPU utilization 85%

Physical host memory utilization 85%

Network throughput 85%

Storage IO latency 20ms

7.2.2 EUE tools information Good EUE is one of the primary factors in determining the success of a VDI implementation. As a result, a

number of vendors have developed toolsets that monitor the environmental parameters that are relevant

to EUE. PAAC on Dell Wyse Datacenter solutions uses the Liquidware Labs Stratusphere UX tool to ensure

that good EUE is delivered for the density numbers defined in our RAs. More specifically, our PAAC analysis

uses a scatter plot provided by Stratusphere UX which presents end-user experience for all load-testing

users. Stratusphere UX does this by algorithmically combining relevant parameters in relation to virtual

machine experience (e.g. login duration) and virtual desktop IO experience (e.g. disk queue length) to

provide a plot that shows end user experience as good, fair or poor using a golden-quadrant type

approach.

7.2.3 EUE real user information To complement the tools-based end user experience information gathered using Stratusphere UX (as

described above) and to provide further certainty around the performance of Dell Wyse Datacenter

solutions, PAAC on our solutions also involves a user logging into one of the solutions when they are fully

loaded (based on the density specified in the relevant RA) and executing user activities that are

representative of the user type being tested (e.g. task, knowledge or power user). An example would be a

knowledge worker executing a number of appropriate activities in Excel. The purpose of this activity is to

verify that the end-user experience is as good as the user would expect on a physical laptop or desktop.

7.2.4 Dell Wyse Datacenter workloads and profiles It is important to understand user workloads and profiles when designing a desktop virtualization solution

in order to understand the density numbers that the solution can support. For our testing, we use three

workload / profile levels, each of which is bound by specific metrics and capabilities. In addition, we use

workloads and profiles that are targeted at graphics-intensive use cases. We have presented detailed

information for these workloads and profiles below, however it is useful to define the terms “workload”

and “profile” as they are used in this document.

Profile – The configuration of the virtual desktop – the number of vCPUs and amount of RAM

configured on the desktop (i.e. visible to the user).

Workload – The set of applications used for performance analysis and characterization (PAAC) of

Dell Wyse Datacenter solutions (e.g. Microsoft Office applications, PDF reader, Internet Explorer

etc.).


7.2.5 Dell Wyse Datacenter profiles The table below presents the profiles used during PAAC of the Dell Wyse Datacenter solutions. These

profiles have been carefully selected to provide the optimal level of resources for the most common use

cases.

Profile Name vCPUs per VM Memory per VM Use Case

Standard 1 2 GB Task worker

Enhanced 2 3 GB Knowledge worker

Professional 2 4 GB Power user

Shared Graphics 2 + shared GPU 3 GB Knowledge worker with high graphics requirements

Dedicated Graphics 4 + dedicated GPU 32 GB Workstation-type user producing complex 3D models

7.2.6 Dell Wyse Datacenter workloads Load testing on each of the profiles described in the above table is carried out using an appropriate

workload that is representative of the relevant use case. In the case of the non-graphics use cases, the

workloads are Login VSI workloads. In the case of graphics use cases, the workloads are specially designed

workloads that stress the VDI environment to a level that is appropriate for the relevant use case. This

information is summarized in the table below.

Profile Name Workload OS Image

Standard Login VSI Light Shared

Enhanced Login VSI Medium Shared

Professional Login VSI Heavy Shared + profile virtualization

Shared Graphics Fishbowl / eDrawings workload Shared + profile virtualization

Dedicated Graphics eDrawings / AutoCAD – SPEC Viewperf workload

Persistent

With respect to the table above, additional information for each of the workloads is given below. It should

be noted that for Login VSI testing, the following login and boot paradigm was used:

For single-server / single-host testing (typically carried out to determine the virtual desktop

capacity of a specific physical server), users were logged in every 30 seconds.

For multi-host / full solution testing, users were logged in over a period of 1-hour, to replicate the

normal login storm in an enterprise environment.

All desktops were fully booted prior to each login attempt.

For all testing, virtual desktops ran an industry-standard anti-virus solution (McAfee VirusScan Enterprise)

in order to replicate a typical customer environment.


7.2.6.1 Login VSI light workload Compared to the Login VSI medium workload described below, the light workload runs fewer applications

(mainly Excel and Internet Explorer with some minimal Word activity) and starts/stops the applications less

frequently. This results in lower CPU, memory and disk IO usage.

7.2.6.2 Login VSI medium workload The Login VSI medium workload is designed to run on 2 vCPUs per desktop VM. This workload emulates a

medium knowledge worker using Office, IE, PDF and Java/FreeMind. The Login VSI medium workload has

the following characteristics

Once a session has been started the workload will repeat (loop) every 48 minutes.

The loop is divided in 4 segments, each consecutive Login VSI user logon will start a different

segments. This ensures that all elements in the workload are equally used throughout the test.

The medium workload opens up to 5 applications simultaneously.

The keyboard type rate is 160 ms for each character.

Approximately 2 minutes of idle time is included to simulate real world users.

Each loop will open and use:

Outlook, browse messages.

Internet Explorer, browsing different webpages and a YouTube style video (480p movie trailer) is

opened three times in every loop.

Word, one instance to measure response time, one instance to review and edit a document.

Doro PDF Printer & Acrobat Reader, the word document is printed and reviewed to PDF.

Excel, a very large randomized sheet is opened.

PowerPoint, a presentation is reviewed and edited.

FreeMind, a Java based Mind Mapping application.

7.2.6.3 Login VSI heavy workload The heavy workload is based on the medium workload except that the heavy workload:

Begins by opening 4 instances of Internet Explorer. These instances stay open throughout the

workload loop.

Begins by opening 2 instances of Adobe Reader. These instances stay open throughout the

workload loop.

There are more PDF printer actions in the workload.

Instead of 480p videos a 720p and a 1080p video are watched.

Increased the time the workload plays a flash game.

The idle time is reduced to 2 minutes.


7.2.7 Workloads running on shared graphics profile Graphics hardware vendors (e.g. NVIDIA) typically market a number of graphics cards that are targeted at

different user segments. Consequently, it is necessary to provide two shared graphics workloads – one for

mid-range cards and the other for high-end cards.

Mid-Range Shared Graphics Workload – The mid-range shared graphics workload is a modified Login VSI

medium workload with 60 seconds of graphics-intensive activity (Microsoft Fishbowl at

http://ie.microsoft.com/testdrive/performance/fishbowl/) added to each loop.

High-End Shared Graphics Workload – The high-end shared graphics workload consists of one desktop

running Heaven Benchmark and n-1 desktops running eDrawings Advanced Animation activity where n =

per-host virtual desktop density being tested at any specific time.

7.2.8 Workloads running on dedicated graphics profile Similarly for pass-through graphics, two workloads have been defined in order to align with graphics cards

of differing capabilities.

Mid-Range Pass-through Graphics Workload – The mid-range pass-through graphics workload consists

of one desktop running Heaven Benchmark and n-1 desktops running eDrawings Advanced Animation

activity where n = per-host virtual desktop density being tested at any specific time.

High-End Pass-through Graphics Workload – One desktop running Viewperf benchmark; n-1 desktops

running AutoCAD auto-rotate activity where n = per host virtual desktop density being tested at any

specific time.

7.3 Testing and validation

7.3.1 Testing process The purpose of the single server testing is to validate the architectural assumptions made around the

server stack. Each user load is tested against 4 runs. A pilot run to validate that the infrastructure is

functioning and valid data can be captured and 3 subsequent runs allowing correlation of data. Summary

of the test results will be listed out in the below mentioned tabular format.

At different stages of the testing the testing team will complete some manual “User Experience” Testing

while the environment is under load. This will involve a team member logging into a session during the run

and completing tasks similar to the User Workload description. While this experience will be subjective, it

will help provide a better understanding of the end user experience of the desktop sessions, particularly

under high load and ensure that the data gathered is reliable.

Login VSI has two modes for launching user’s sessions:

Parallel – Sessions are launched from multiple launcher hosts in a round robin fashion; this mode is

recommended by Login Consultants when running tests against multiple host servers. In parallel mode the

http://ie.microsoft.com/testdrive/performance/fishbowl/


VSI console is configured to launch a number of sessions over a specified time period (specified in

seconds)

Sequential - Sessions are launched from each launcher host in sequence; sessions are only started from a

second host once all sessions have been launched on the first host- this is repeated for each launcher

host. Sequential launching is recommended by Login Consultants when testing a single desktop host

server. The VSI console is configured to launch a specific number of sessions at a specified interval

specified in seconds

All test runs were conducted using the Login VSI “Parallel Launch” mode, all sessions were launched over

an hour to try and represent the typical 9am logon storm. Once the last user session has connected, the

sessions are left to run for 15 minutes prior to the sessions being instructed to logout at the end of the

current task sequence, this allows every user to complete a minimum of two task sequences within the

run before logging out. The single server test runs were configured to launch user sessions every 60

seconds, as with the full bundle test runs sessions were left to run for 15 minutes after the last user

connected prior to the sessions being instructed to log out.

7.4 VMware Horizon View test results

7.4.1 Configuration Validation for this project was completed for VMware View 6.1 on the following platforms.

ESXi 6.0 / VMware View 6.1.1 / R730

VMware View 6.1.1 was used to provision the user desktops. The desktops were non-persistent linked

clone desktops. The desktops were captured from a Windows 8.1 master image.

Platform configurations are shown below and the Login VSI workloads used for load testing on each

environment.

Host CPU Memory RAID Ctlr HDD Config Network

R730 (compute)

E5-2697v3 (14 Core, 2.6GHz)

384GB DDR4

H730P Mini Embedded

10 x 300GB SAS 2.5” (T1)

Broadcom 57800 Gigabit

R720 (mgmt)

E5-2690 (8 Core, 2.9 GHz)

384GB DDR3

H710P Mini Embedded

10 X 146GB SAS 2.5” (T1)

Broadcom 5720 Gigabit

1GB networking was used for both configurations.

Compute and Management resources were split out with the following configuration and all test runs

were completed with this configuration.

Node 1 – R720– Dedicated Management (vCenter Appliance 6.0, SQL Server, VMware View

Connection Server 6.1.1, VMware View Composer 6.1.1)

Node 2 – R730– Dedicated Compute Host.


The virtual machines were non-persistent linked clone desktops each configured on Windows 8.1 aligning

with the Login VSI 4.X virtual machine configuration, Office 2010 was used with each Virtual Machine sized

at 32 GB. User Workload configuration of the load generation virtual machines is shown in the table

below.

User Workload vCPUs Memory OS Bit Level HDD Size

Standard User 1 2 GB x32 24 GB

Enhanced User 2 3 GB x32 24 GB

Professional User 2 4 GB x64 32 GB

As a result of the testing, the following density numbers can be applied to the individual solutions. In all

cases CPU percentage used was the limiting factor. Memory usage, IOPs and network usage were not

strained.

The following table summarizes the user workload resources and densities as tested:

Workload VM Density Desktop Type

Standard 230 Linked Clone

Enhanced 170 Linked Clone

Professional 130 Linked Clone

Windows 7 and 8 VMware Horizon View best practices for optimizing desktops were followed. Details for

these are located here: http://www.vmware.com/resources/techresources/10157

Windows 8.1 desktops were configured with some optimizations to enable the Login VSI workload to run

and in order to prevent long delays in the login process. Previous experience with Windows 8.1 has shown

that the login delays are somewhat longer that experienced with Windows 7. These were alleviated by

performing the following customizations

Bypass Windows Metro screen to go straight to the Windows Desktop. This is performed by a

scheduled task provided by Login Consultants at logon time.

Disable the “Hi, while we’re getting things ready…” first time login animation. In randomly assigned

Desktop groups each login is seen as a first time login. This registry setting can prevent the

animation and therefore the overhead associated with it.

[HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\Policies\System]

"EnableFirstLogonAnimation"=dword:00000000

McAfee antivirus is configured to treat the Login VSI process VSI32.exe as a low risk process and to

not scan that process. Long delays during login of up to 1 minute were detected as VSI32.exe was

scanned.

Before finalizing the Golden template image, perform a number of logins using domain accounts.

This was observed to significantly speed up the logon process for VMs deployed from the Golden

http://www.vmware.com/resources/techresources/10157


image. It is assumed that Windows 8 has a learning process when logging on to a domain for the

first time.

7.4.2 ESXi 6.0/View 6.1

7.4.2.1 Standard user workload (230 users) For this testing run the R730 compute host was populated with 230 non-persistent, linked clone virtual

machines provisioned by VMware View 6.1.1.

This chart includes the additional 21% of CPU available from the Turbo boost feature. Without the

inclusion there is a total of 72,800 MHz available for desktops, with Turbo boost the total available MHz

value is 88,088 MHz.

The CPU reaches a steady state average of 96% during the test cycle when approximately 230 users are

logged on and a maximum of 97%.

230 user graphs:

0

20

40

60

80

100

120

140

12

:40

12

:50

13

:00

13

:10

13

:20

13

:30

13

:40

13

:50

14

:00

14

:10

14

:20

14

:30

14

:40

14

:50

15

:00

CPU Utilization

CPU Usage

CPU Threshold 85%

Turbo Performance Increase21%


With regard to host memory consumption, out of a total of 384 GB available memory there were no

constraints on the host even though consumed memory was quite high. The compute host reached a max

memory consumption of 375 GB with active memory usage reaching a max of 132 GB. There was some

memory ballooning followed by memory swapping as the number of logged on desktops increased

towards the latter stages of testing.

Network bandwidth is not an issue on this solution with a steady state peak of approximately 29,000 Kbps.

310

320

330

340

350

360

370

380

12

:40

12

:50

13

:00

13

:10

13

:20

13

:30

13

:40

13

:50

14

:00

14

:10

14

:20

14

:30

14

:40

14

:50

15

:00

Consumed Memory

Consumed GB

0

20

40

60

80

100

120

140

12

:40

12

:50

13

:00

13

:10

13

:20

13

:30

13

:40

13

:50

14

:00

14

:10

14

:20

14

:30

14

:40

14

:50

15

:00

Active Memory

Active GB


The Login VSI Max user experience score for this test indicates that the VSI Max score reached after

approximately 190 users were logged on. There was little observed deterioration of user experience during

testing as mouse and window response both remained good.

7.4.2.2 Enhanced user workload (170 users) For this testing run the R730 compute host was populated with 170 non-persistent, linked clone virtual

machines provisioned by VMware View 6.1.

0

5000

10000

15000

20000

25000

30000

35000

12

:40

12

:50

13

:00

13

:10

13

:20

13

:30

13

:40

13

:50

14

:00

14

:10

14

:20

14

:30

14

:40

14

:50

15

:00

Network Kbps

Network Kbps







170 user graphs:

With regard to host memory consumption, with 384 GB of physical memory, the compute host reached a

max memory consumption of 316 GB with active memory usage reaching a max of 133 GB. There was

some memory ballooning towards the end of the test run but no memory swapping took place.

0

20

40

60

80

100

120

140

12

:30

12

:40

12

:50

13

:00

13

:10

13

:20

13

:30

13

:40

13

:50

14

:00

14

:10

14

:20

CPU Utilization

CPU Usage

CPU Threshold 85%


0

50

100

150

200

250

300

350

400

12

:30

12

:35

12

:40

12

:45

12

:50

12

:55

13

:00

13

:05

13

:10

13

:15

13

:20

13

:25

13

:30

13

:35

13

:40

13

:45

13

:50

13

:55

14

:00

14

:05

14

:10

14

:15

14

:20

Consumed Memory

Consumed GB



The Login VSI Max user experience score for this test indicates that the VSI Max score reached but did not

go much beyond the threshold until near the end of the test cycle. There was little observed deterioration

of user experience during testing as video playback and mouse response both remained good.

0

20

40

60

80

100

120

140

160

12

:30

12

:35

12

:40

12

:45

12

:50

12

:55

13

:00

13

:05

13

:10

13

:15

13

:20

13

:25

13

:30

13

:35

13

:40

13

:45

13

:50

13

:55

14

:00

14

:05

14

:10

14

:15

14

:20

Active Memory

Active GB

0

5000

10000

15000

20000

25000

30000

35000

12

:30

12

:35

12

:40

12

:45

12

:50

12

:55

13

:00

13

:05

13

:10

13

:15

13

:20

13

:25

13

:30

13

:35

13

:40

13

:45

13

:50

13

:55

14

:00

14

:05

14

:10

14

:15

14

:20

Network Kbps

Network Kbps


7.4.2.3 Professional user workload (130 users) For this testing run the R730 compute host was populated with 130 non-persistent, linked clone virtual

machines provisioned by VMware View 6.1.1.






130 User graphs:

0

20

40

60

80

100

120

140

16

:50

16

:55

17

:00

17

:05

17

:10

17

:15

17

:20

17

:25

17

:30

17

:35

17

:40

17

:45

17

:50

17

:55

18

:00

18

:05

18

:10

18

:15

18

:20

18

:25

18

:30

CPU Utilization

CPU Usage

CPU Threshold 85%



With regard to host memory consumption, out of a total of 384 GB available memory there were no

constraints. The Compute host reached a max memory consumption of 325 GB with active memory

usage reaching a max of 83 GB. There was no memory ballooning and swapping.


0

50

100

150

200

250

300

350

16

:50

16

:55

17

:00

17

:05

17

:10

17

:15

17

:20

17

:25

17

:30

17

:35

17

:40

17

:45

17

:50

17

:55

18

:00

18

:05

18

:10

18

:15

18

:20

18

:25

18

:30

Consumed Memory

Consumed GB

0

10

20

30

40

50

60

70

80

90

16

:50

16

:55

17

:00

17

:05

17

:10

17

:15

17

:20

17

:25

17

:30

17

:35

17

:40

17

:45

17

:50

17

:55

18

:00

18

:05

18

:10

18

:15

18

:20

18

:25

18

:30

Active Memory

Active GB


The Login VSI Max user experience score for this test indicates that the VSI Max score was reached close

to the end of the test cycle indicating there was little deterioration of user experience during testing. This

is also borne out by the fact that video playback and mouse and window response was good even at the

end of the test cycle.

Notes:

As indicated above, the CPU graphs do not take into account the extra 21% of CPU resources

available through the 2697v3’s turbo feature.

0

5000

10000

15000

20000

25000

30000

16

:50

16

:55

17

:00

17

:05

17

:10

17

:15

17

:20

17

:25

17

:30

17

:35

17

:40

17

:45

17

:50

17

:55

18

:00

18

:05

18

:10

18

:15

18

:20

18

:25

18

:30

Network Kbps

Network Kbps


Subjective user experience showed mouse movement and window response times when clicking

within a running session during steady state were good. Video playback was good on the

Professional and Enhanced workloads even with all desktops logged on.

User login times were consistent around the 21 - 24 second mark for the Professional workload,

30 – 33 seconds for the Enhanced workload and 28 -30 seconds on the Standard workload. A few

session took some extra time to login on all workloads especially towards the end of the test

cycles.

384GB of memory saw some ballooning and swapping while running the Standard workload

testing, some ballooning only on the Enhanced workload but none on the Professional. The

greater number of desktops on the host causes more ballooning and swapping.

7.5 Dell PowerEdge C4130 testing

7.5.1 Configuration Validation for this project was completed for VMware Horizon View 6.1 on the following platform:

PowerEdge C4130 / VMware ESXi 6.0 / VMware Horizon View 6.1.0

VMware Horizon View was used to provide the persistent, full-clone user desktops. The desktops were

stored locally and were created from a Windows 7 master image. The View Storage Accelerator was

disabled for the host and the desktop virtual machines.

Access to the virtual desktops for test purposes was provided through Windows 7 virtual machines loaded

with Windows 7 and View Client 3.3.0.

Platform configurations are shown below and the Login VSI workloads used for load testing on each

environment.

Host CPU Memory RAID Ctlr HDD Config Network

PE C4130 E5-2670 v3 (12 Core, 2.3GHz)

512 GB N/A 2 x 800GB SATA 1.8” (T1)

Intel 10GbE

10 GbE networking was used for the tests. Four NVidia GRID K2 GPU cards were used for all benchmarks.

Compute and Management resources were split out with the following configuration and all test runs

were completed with this configuration.

Node 1 – R920– Dedicated Management

Node 2 – C4130– Dedicated Compute Host

The virtual machines were full clone desktops each configured with Windows 7. SPECwpc 1.2 was used to

generate a graphics workload based upon Dassault Systemes Solidworks. User Workload configuration of

the load generation virtual machines is shown in the table below.


User Workload vCPUs Memory OS Bit Level

HDD Size vGPU Profile Graphics Memory

Custom (SPECwpc sw-03)

3 16 GB x64 120 GB K280Q 4 GB

Custom (SPECwpc sw-03)

3 16 GB x64 120 GB K260Q 2 GB

7.5.2 Test results Validation was performed using CCC testing methodology using LoginVSI 4 load generation tool for VDI

benchmarking that simulates production user workloads.

Each test run adhered to PAAC best practices with a 30 second session logon interval followed by 1 hour

of steady state after which sessions would begin logging off.

The following table summarizes the test results for the workload once it achieved steady state.

Hypervisor vGPU Profile

Workload VMs per

host

Avg. CPU

%

Avg. Memory

Active

Avg. GPU %

Avg. GPU Memory %

Avg. Net MB/s/User

ESXi 6.0 K280Q Custom 8 56% 34 GB 30.03% 10.56% 2.7 MB/sec

ESXi 6.0 K260Q Custom 16 93% 75 GB 50.40% 17.12% 1.4 MB/sec

CPU Utilization - CPU % for ESX Hosts was adjusted to account for the fact that on Intel E5-2670v3 series

processors the ESX host CPU metrics will exceed the rated 100% for the host if Turbo Boost is enabled (by

default). The Adjusted CPU % Usage is based on 100% usage and but is not reflected in the charts. The

figure shown in the table is the Compute host steady state peak CPU Usage.

Memory Utilization - The figure shown in the table above is the average memory consumed per Compute

host over the recorded test period. Active is the average active memory per Compute host over the

recorded test period.

Network Utilization - The figure shown in the table is the average MB/sec per user over the recorded test

period.

7.5.2.1 K280Q vGPU profile, 8 user test For this testing run, the C4130 compute host was populated with 8 full clone virtual machines with the

K280Q vGPU profile provisioned by VMware Horizon View 6.1.

This chart does not include the additional CPU available from the Turbo boost feature.

The CPU reached a steady state peak of 56.94% during the test cycle when 8 users are logged on.


In regards to memory consumption for the host, out of a total of 512 GB available memory there were no

constraints on the host. The compute host reached a max memory consumption of 137 GB with active

memory usage reaching a max of 39 GB. There was no memory ballooning and swapping.

0

10

20

30

40

50

60

14:25 14:30 14:35 14:40 14:45 14:50 14:55 15:00 15:05 15:10 15:15 15:20 15:25

Pe

rce

nt

Time

Poweredge C4130 CPU Utilization

137.915

137.92

137.925

137.93

137.935

137.94

137.945

137.95

137.955

137.96

137.965

14:25 14:30 14:35 14:40 14:45 14:50 14:55 15:00 15:05 15:10 15:15 15:20 15:25

Consumed Memory

Consumed GB



0

5

10

15

20

25

30

35

40

45

14:25 14:30 14:35 14:40 14:45 14:50 14:55 15:00 15:05 15:10 15:15 15:20 15:25

Active Memory

Active GB

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

14:25 14:30 14:35 14:40 14:45 14:50 14:55 15:00 15:05 15:10 15:15 15:20 15:25

Memory Swap And Ballooning

Swap used

Balloon


GPU utilization was also not an issue with a steady state peak of approximately 35%.

7.5.2.2 K260Q vGPU profile, 16 user test For this testing run, the C4130 compute host was populated with 16 full clone virtual machines with the

K260Q vGPU profile provisioned by VMware Horizon View 6.1.

This chart does not include the additional CPU available from the Turbo boost feature.

The CPU reached a steady state peak of 93% during the test cycle when 16 users are logged on.

0

5000

10000

15000

20000

25000

30000

14:25 14:30 14:35 14:40 14:45 14:50 14:55 15:00 15:05 15:10 15:15 15:20 15:25

Network KBps

0.00%

5.00%

10.00%

15.00%

20.00%

25.00%

30.00%

35.00%

40.00%

1 2 3 4 5 6 7 8 9 10 11 12 13

Pe

rce

nt

5 minute intervals

C4130/K280Q GPU Utilization

GPU1 Utilization (%)









In regards to memory consumption for the host, out of a total of 512 GB available memory there were no

constraints on the host. The compute host reached a max memory consumption of 269 GB with active

memory usage reaching a max of 78 GB. There was no memory ballooning and swapping. There is a

large amount of active memory at the beginning of the test as the VMs were rebooted in between tests

and not enough time was allowed for the active memory to be released.

0

10

20

30

40

50

60

70

80

90

100

14:00 14:05 14:10 14:15 14:20 14:25 14:30 14:35 14:40 14:45 14:50

Pe

rce

nt

Time

Poweredge C4130 CPU Utilization

268.3

268.35

268.4

268.45

268.5

268.55

268.6

268.65

268.7

268.75

14:00 14:05 14:10 14:15 14:20 14:25 14:30 14:35 14:40 14:45 14:50 14:55 15:00 15:05 15:10

Consumed GB

Consumed GB



0

50

100

150

200

250

14:00 14:05 14:10 14:15 14:20 14:25 14:30 14:35 14:40 14:45 14:50 14:55 15:00 15:05 15:10

Active GB

Active GB

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

14:00 14:05 14:10 14:15 14:20 14:25 14:30 14:35 14:40 14:45 14:50 14:55 15:00 15:05 15:10

Memory Swap and Balooning

Swap used

Balloon


GPU utilization was also not an issue with a steady state peak of approximately 52%.

7.6 Dell EqualLogic PS6210XS testing with VMware Horizon View

7.6.1 Overview The objective of this testing was to demonstrate 2000 standard (now called enhanced workloads) users

would perform through various states of the environment. A single PS6210XS was leveraged for this test.

The test infrastructure used the following:

0

5000

10000

15000

20000

25000

30000

35000

40000

45000

50000

14:00 14:05 14:10 14:15 14:20 14:25 14:30 14:35 14:40 14:45 14:50 14:55 15:00 15:05 15:10 15:15 15:20 15:25

Network KBps

0.00%

10.00%

20.00%

30.00%

40.00%

50.00%

60.00%

70.00%

1 2 3 4 5 6 7 8 9 10 11 12 13 14

Pe

rce

nt

5 minute intervals

Poweredge C4130/K260Q GPU Utilization










VMware Horizon View 5.2 (latest available at the time of this test)

VMware vSphere 5.1 (latest available at the time of this test to support View 5.2)

Dell PowerEdge R730 (4) and M620 Servers (16) not including additional PowerEdge Servers to

support VDI Load generation services.

Dell Force10 and PowerConnect switches

Dell EqualLogic storage arrays (PS6210XS and PS6510E)

During this 2000 VDI desktop test the PS6210XS generated which included satisfactory performance

across the entire VDI infrastructure:

Boot Storm IOPS Login Storm IOPS Steady State IOPS ms Avg. Latency

17514 17144 16773 5

7.6.2 Compute resources

The entire infrastructure and test configuration was installed in a single Dell PowerEdge M1000e blade

chassis complete with 16 PowerEdge M620 blade servers and four additional PowerEdge R730 rack

servers. The ESXi clusters used include:

Infrastructure Cluster: PowerEdge M620 blade server hosting virtual machines for Active Directory

services, VMware vCenter 5.1 server, Horizon View 5.2 server (primary and secondary), Horizon

View Composer server, Microsoft™ Windows Server™ 2008 R2 based file server and SQL Server®

2008 R2.

Horizon View Client Clusters: Four PowerEdge R730 rack servers and 15 PowerEdge M620 blade

servers hosting virtual desktops.

7.6.3 Network resources

The following considerations were made for designing the network of the VDI solution presented in this

reference architecture:

Two PowerConnect M8024-K blade switches in Fabric A for connectivity to the dedicated iSCSI

SAN

Two PowerConnect M6348 blade switches stacked in Fabric B for connectivity to the

Management LAN, VDI client LAN and a vMotion LAN.

Each PowerEdge M620 blade server configured with one Broadcom 57810S Dual Port 10 GbE NIC

card and one Broadcom 5719 Quad Port 1 GbE NIC Card. Broadcom 57810S card was assigned as

Fabric A LOM and the other Broadcom 5719 1Gb NIC card was assigned as Fabric B on the blade

chassis.

Fabric A was a 10 GbE network dedicated for iSCSI traffic while Fabric B carried the VDI traffic for

all 2,000 VMs.


The PowerConnect switches in Fabric B were interconnected using the stacking modules to

provide high availability and redundancy of the VDI fabric.

Fabric C was unused.

Two Force10 S4810 switches were used for external SAN access. These switches were stacked

together for failure resiliency and ease of management.

7.6.4 iSCSI SAN configuration overview

In the figure below it shows the network connectivity between a single PowerEdge M620 Server and the storage array though the blade server chassis:

7.6.5 Test objectives: Determine how many virtual desktops can be deployed in a Horizon View environment using a

single EqualLogic PS6210XS storage array with acceptable user experience indicators for a

standard user workload profile.

Determine the performance impact on the storage array of peak I/O activity such as boot and

login storms

Develop sizing guidelines for Horizon View VDI deployments leveraging EqualLogic PS6210XS

hybrid storage arrays


The “medium” workload from LoginVSI was used to simulate desktop workloads for each of the

2000 desktops. LoginVSI 3.7 was used.

7.6.6 Test criteria/thresholds: Maintain less than 20ms storage disk latency on average

CPU utilization on any ESXi server should not exceed 85%

No memory ballooning on any desktop VM

Total network bandwidth utilization should not exceed 90% on any one link

TCP/IP storage network retransmissions should be less than 0.5%

Stratusphere UX’s scatterplot should report desktops in the acceptable user experience range.

7.6.7 Boot storm I/O To simulate a boot storm, the virtual desktops were reset simultaneously from the VMware vSphere client.

The figure below shows the storage characteristics during the boot storm – the EqualLogic PS6210XS

array delivered 17,514 IOPS (8.7 IOPS per VM) with less than 3 ms average latency under the peak load

during this test and all 2,000 desktops were available in about 25 minutes.

The above screenshot represents the boot storm for 2,000 VMware Linked Clone VMs with a read/write

ratio of 65% read and 35% write. The Replica volumes contributed to the majority of the I/O. Latency was

extremely low at a weighted average of 2.78 ms.

Due to the large number of VMs being powered on, each Replica volume generated its individual

maximum IOPS at different times during the boot storm, depending on when VMs on a particular Replica

volume got powered on. The next figure below shows two Replica volumes generating the majority of


IOPS when the boot storm itself was at its peak. I/O on Replica volumes were virtually 100% read

operations.

Storage network utilization was well within the available bandwidth. The peak network utilization during

the boot storm reached approximately 6% of the total storage network bandwidth and then gradually

declined once all the VMs were booted up. There were also no retransmissions on the iSCSI SAN.

The above results show that EqualLogic PS6210XS hybrid array can handle a heavy I/O load like a boot

storms in a VDI environment with no issues.

7.6.8 Login storm I/O Login VSI was configured to launch 2,000 virtual desktops over a period of about 30 minutes after pre-

booting the virtual desktops. The peak IOPS observed during the login storm was about 17,144 IOPS (8.5

IOPS per VM).

Login storms generate significantly more write IOPS than a boot storm due to multiple factors including:

• User profile activity

• Starting operating system services on the desktop

• First launch of applications


Once a virtual desktop has achieved a steady state after user login, the Windows 7 OS has cached

applications in memory and does not need to access storage each time the application is launched. This

leads to lower IOPS during the steady state. The figure below shows the IOPS and latency observed during

the login storm.

The EqualLogic PS6210XS easily handled logging in 2,000 sessions in a short time delivering the required

17,144 IOPS with 4.2 ms of average latency at the peak of the login storm. Table 7 shows the overall disk

usage in the array during the login storm.

Most of the login storm I/O operations are handled by SSDs and therefore the array is able to provide the

best possible performance. Each SSD handled approximately 4,950 IOPS at the peak of login storm; the


average latency was very low during the entire login storm time period and the array clearly demonstrated

its ability to handle the workload.

7.6.9 Steady state I/O Following the completion of the login storm, the I/O profile changed to approximately 24% Read and 76%

Write I/O operations at steady state. The total IOPS required during the peak load at steady state with all

the users logged in was around 16,773 (8.3 IOPS per VM). EqualLogic PS6210XS delivered these IOPS with

5 ms average latency, which is well below the 20 ms threshold. The average load during the entire steady

26 Desktop Virtualization with VMware Horizon View 5.2 on Dell EqualLogic PS6210XS Hybrid Storage

Array state test period was approximately 15,000 IOPS, which EqualLogic PS6210XS delivered with 4.7 ms

average latency as shown in Figure below.

All changes that occur on the virtual desktop (including temporary OS writes such as memory paging) are

being written to disk. The I/O pattern is mostly writes due to this activity. Once the desktops are booted

and in a steady state, the read I/O becomes minimal due to Horizon View Storage Accelerator enabling

content based read caching (CBRC) on the ESXi hosts.

During steady state there is minimal activity on the replica volume and most of the activity is seen on the

VDI-Images volumes that host the virtual desktops.

The figure below shows the performance of the array during the steady state test.



7.6.10 Server host performance All compute infrastructure values were performance thresholds as previously described and shown below:


7.6.11 Summary Following are the key observations made over the course of validation:

A single EqualLogic PS6210XS was able to host 2,000 virtual desktops and support a standard user

type of I/O activity.

The VDI I/O was mostly write-intensive I/O with more than 74% writes and less than 26% reads.


None of the system resources on the ESXi servers hosting the virtual desktops reached maximum

utilization levels at any time.

During the boot storm simulation, nearly 17,500 IOPS with less than 2.8 ms of average latency

were observed and all the 2,000 desktops were available in Horizon View within 25 minutes of the

storm.

To simulate a login storm, 2,000 users were logged in within a span of 30 minutes. A single

EqualLogic PS6210XS array was able to easily sustain this login storm with approximately 17,150

IOPS and 4.2 ms average. Most of the I/O was served by the SSDs on the array.

The user experience for 2,000 desktops was well within acceptable limits. All sessions were in the

upper right quadrant and virtually all of them were in the Good category on Stratusphere UX

scatter plot.

For full results and information please see this document http://dell.to/1g4Gc9v.

http://dell.to/1g4Gc9v


Acknowledgements

Thanks to the Darin Schmitz and Damon Zaylskie of the Dell Compellent MSV Solutions team for providing

expertise and validation of the Dell Wyse Datacenter Compellent Tier 1 array.

Thanks to Paul Wynne and the Dell Wyse Solutions Ingredients Extended Team for their expertise and

continued support validating VDI architectures and Tier 1 storage.

Thanks to Sujit Somandepalli and Chhandomay Mandal of Storage Engineering and Storage Technical

Marketing respectively for Storage contributions.

Thanks to John Kelly of the Dell Wyse Solutions Engineering team for his expertise and guidance in the

Dell Wyse Datacenter PAAC process.

About the authors

Gus Chavira is the Senior Principal Engineering Architect for VMware Horizon based solutions at Dell. Gus

has extensive experience and expertise on the VMware solutions software stacks as well as in Enterprise

virtualization, storage and enterprise data center design. Gus has worked in capacities of Sys Admin, DBA,

Network and Storage Admin, Virtualization Practice Architect, Enterprise and Solutions Architect. In

addition, Gus carries a B.S. in Computer Science

Peter Fine is the Senior Principal Engineering Architect for VDI-based solutions at Dell. Peter has extensive

experience and expertise on the broader Microsoft, Citrix and VMware solutions software stacks as well as

in enterprise virtualization, storage, networking and enterprise data center design.

Andrew McDaniel is the Solutions Development Manager for VMware solutions at Dell, managing the

development and delivery of enterprise-class desktop virtualization solutions based on Dell Data center

components and core virtualization platforms.

Nicholas Busick is a Senior Solutions Engineer with Dell Wyse Solutions Engineering building, testing,

validating and optimizing enterprise VDI stacks.

Darpan Patel is a Senior Solutions Engineer with Dell Wyse Solutions Engineering with extensive

experience in validating, building and optimizing enterprise class VDI solutions on Microsoft (Hyper-V),

VMware (View) and Citrix (XenDesktop). Darpan has a master’s degree in Information Systems from Pace

University in New York and is VCP5-DCV certified. (VMware Certified Professional 5 – Data Center

Virtualization).

David Hulama is a Senior Technical Marketing Advisor for VMware Horizon View solutions at Dell. David

has a broad technical background in a variety of technical areas and expertise in enterprise-class

virtualization solutions.

Documents

Dell Wyse Datacenter for VMware Horizon View … 2014 Updated to include 13g ... 3 Dell Wyse Datacenter for VMware Horizon View Reference Architecture ... Dell Wyse Datacenter for