Technical data

Solution Architectural Overview

VMware Horizon View 5.3 and VMware vSphere for up to 2,000 Virtual

Desktops Enabled by Brocade Network Fabrics, EMC VNX, and EMC Next-

Generation Backup

system is provisioned from the pool to present to the vSphere

servers as a VMFS or NFS datastore.

 Thirty-two NL-SAS disks (shown as 0_0_8, 1_0_3, 1_1_0 to 1_1_14,

and 0_2_0 to 0_2_14) in the RAID 6 Storage Pool 1 are used to

store user data and profiles. FAST Cache is enabled for the entire

pool. Ten LUNs of 3 TB each are provisioned from the pool to

provide the storage required to create four CIFS file systems.

 Disks shown as 1_2_10 to 1_2_14 are unbound and not used for

testing this solution.

 Disks shaded gray are required and are part of the core storage

layout.

If multiple drive types have been implemented, FAST VP may be enabled

to automatically tier data to balance differences in performance and

capacity.

VNX shared file systems

Four shared file systems are used by the virtual desktops—two for the

VMware View Persona Management repositories and two to redirect user

storage that resides in home directories. In general, redirecting users’ data

out of the base image to VNX for File enables centralized administration,

backup and recovery, and makes the desktops more stateless. Each file

system is exported to the environment through a CIFS share. Each persona

management repository share and home directory share serves 1,000

users.

High availability and failover

This VSPEX solution provides a highly available virtualized server, network,

and storage infrastructure. When implemented in accordance with this

guide it provides the ability to survive single-unit failures with minimal

impact to business operations.

EMC recommends configuring high availability in the virtualization layer

and automatically allowing the hypervisor to restart virtual machines that

fail. Figure 29 illustrates the hypervisor layer responding to a failure in the

compute layer.

Figure 29. High availability at the virtualization layer

By implementing high availability at the virtualization layer, even in the

event of a hardware failure, the infrastructure will attempt to keep as

many services running as possible.

Introduction

Virtualization

layer