VMware Horizon 6.

0 with View
Reference Architecture

V1.0 August 2014

Copyright 2014 Nutanix, Inc.

All rights reserved. This product is protected by U.S. and
international copyright and intellectual property laws.
Nutanix is a trademark of Nutanix, Inc. in the United States
and/or other jurisdictions. All other marks and names
mentioned herein may be trademarks of their respective
companies.

VMware Horizon 6.0 with View | 2

Table of Contents
1.

Executive Summary .......................................................................................................... 4

2. Audience and Purpose ..................................................................................................... 5

3. Solution Overview ............................................................................................................. 6

Web-scale Powering Desktops on Demand .................................................................... 6
Nutanix Architecture ................................................................................................................. 7
Better together: Nutanix and VMware ............................................................................. 10
VMware Horizon 6.0 with View ........................................................................................... 12

4. Solution Design ................................................................................................................. 13

Desktop & RDS Sizing ............................................................................................................ 16
Desktop and RDS Optimizations......................................................................................... 17
Horizon View Composer.........................................................................................................18
View Composer with Shadow Clones and VCAI .......................................................... 19
Shadow Clones .......................................................................................................................... 19
Nutanix Support for View Composer API for Array Integration ........................... 20
Nutanix Web-scale Converged Infrastructure ............................................................... 21
Network ........................................................................................................................................ 21
Logical Network Design ........................................................................................................ 22

5. Validation and Benchmarking .................................................................................... 24

Nutanix Configuration ........................................................................................................... 24
Login VSI Benchmark ............................................................................................................ 26

6. Validation Results ........................................................................................................... 30

View Composer with VCAI for a 4-node NX-3060 (NX-3460) ............................ 30
View Composer with VCAI for a 4-node NX-3060 (NX-3460+NX-3260) ........ 32
View Composer with VCAI with 8-nodes NX-3060 (2x NX-3460) ..................... 34
Linear Scale ................................................................................................................................ 37
Boot Storm ................................................................................................................................. 37
High Availability and Continuity .........................................................................................38
Remote Desktop Services on NX-3460 (4-nodes) .................................................... 39

7. Solution Application ...................................................................................................... 42

Scenario: 12 Nodes .................................................................................................................. 42
Scenario: 24 Nodes................................................................................................................. 44

8. Conclusion ......................................................................................................................... 47
9. Appendix: Configuration .............................................................................................. 48
10.

References....................................................................................................................... 49
Table of Figures ....................................................................................................................... 49
Table of Tables ......................................................................................................................... 50

11.

About the Authors ....................................................................................................... 52

VMware Horizon 6.0 with View | 3

1. Executive Summary
This validated reference architecture document highlights the use of Nutanix webscale converged infrastructure to seamlessly scale and deliver consistent robust
performance for VMware Horizon 6 (with View). Nutanix is able to eliminate
bottlenecks and deliver both desktops and RDS (Remote Desktops Services) across
many nodes to meet small or large business requirements. This paper demonstrates
scaling from 4 to 6 to 8 nodes.

The following table provides highlights of the testing results:

Linear scaling w ith over 110 m edium (2 vCPU)
w orkload desktops per node supports 440 users
in 2U, inclusive of com pute and storage.
Room to grow -- VSIm ax w as not reached at 886
users w ith 8 Nutanix nodes

Consistent desktop perform ance at any scale

O nly 6 m inutes to boot over 880 desktops

Zero dow ntim e during loss of storage controller

View Com poser Array Integration for fast nonim pacting cloning

Clone large 100+ GB RDS servers in 8 seconds

VMware Horizon 6.0 with View | 4

2. Audience and Purpose

This reference architecture document is part of the Nutanix Solutions Library. It is
intended for architects and systems engineers responsible for designing, managing,
and supporting Nutanix infrastructures running VMware Horizon with View.
Consumers of this document should already be familiar with vSphere, Horizon with
View, and Nutanix.

This document will cover the following subject areas:

o

Overview of the Nutanix solution

Overview of VMware Horizon with View and its use cases

The benefits of VMware Horizon with View on Nutanix

Architecting a complete VMware Horizon with View on the Nutanix platform

Sizing guidance for scaling VMware Horizon with View deployments on

Nutanix

Design and configuration considerations when architecting a VMware Horizon

with View solution on Nutanix

Benchmarking VMware Horizon with View performance on Nutanix using

both Windows 7 for VDI and Windows 2012 R2 for Remote Desktop Services
(RDS)

VMware Horizon 6.0 with View | 5

3. Solution Overview
Web-scale Powering Desktops on Demand
Nutanix delivers an out-of-the-box infrastructure solution for virtual desktops that
eliminates the high cost, variable performance, and extensive risk of conventional
solutions. The Nutanix web-scale converged infrastructure is a turnkey solution that
comes ready to run VMware Horizon View. The Nutanix platforms unique
architecture allows enterprises to scale their virtual desktops from 50 to tens of
thousands of desktops in a linear fashion, providing customers with a simple path to
enterprise deployment with the agility of public cloud providers.

Figure 1 Web-scale properties of Nutanix

The Nutanix platform supports every type of VDI user, from task and knowledge
workers to power and data scientists. Whether you have persistent desktops that
are customized for knowledge workers, shared hosted virtual desktops (HVD) for a
general workforce, or the most 3D graphics intensive users, Nutanix provides the
right resources in a single-box solution.

VMware Horizon 6.0 with View | 6

Nutanix Architecture
The Nutanix web-scale converged infrastructure is a scale-out cluster of highperformance nodes (or servers), each running a standard hypervisor and containing
processors, memory, and local storage (consisting of SSD Flash and high capacity
SATA disk drives). Each node runs virtual machines just like a standard virtual
machine host.

Figure 2 Nutanix Node Architecture

In addition, local storage from all nodes is virtualized into a unified pool by the
Nutanix Distributed File System (NDFS). In effect, NDFS acts like an advanced NAS
that uses local SSDs and disks from all nodes to store virtual machine data. Virtual
machines running on the cluster write data to NDFS as if they were writing to shared
storage.

Figure 3 Nutanix Architecture

VMware Horizon 6.0 with View | 7

NDFS understand the concept of a virtual machine and provides advanced data
management features. It brings data closer to virtual machines by storing the data
locally on the system, resulting in higher performance at a lower cost. Nutanix
platforms can horizontally scale from as few as three nodes to a large number of
nodes, enabling organizations to scale their infrastructure as their needs grow.
The Nutanix Elastic Deduplication Engine is a software-driven, massively scalable
and intelligent data reduction technology. It increases the effective capacity in the
disk tier, as well as the RAM and flash cache tiers of the system, by eliminating
duplicate data. This substantially increases storage efficiency, while also improving
performance due to larger effective cache capacity in RAM and flash. Deduplication
is performed by each node individually in the cluster, allowing for efficient and
uniform deduplication at scale. This technology is increasingly effective with
full/persistent clones or P2V migrations.
Sequential streams of data are fingerprinted
at 4K granularity for efficient deduplication
VM
VM 11

...

VM
VM 11

Hypervisor
Hypervisor

Cache

...

VM
VM N
N

CVM
CVM

Cache

VM
VM 11

...

Hypervisor
Hypervisor

Cache
Cache
Storage
Storage

Each node participates in, and performs,

its own fingerprinting and deduplication

Only a single instance of the shared VM

data is pulled into the cache upon read

VM
VM N
N

Cache
Cache
CVM
CVM

Storage
Storage

Cache

...

VM
VM N
N

Hypervisor
Hypervisor
Cache
Cache
Storage
Storage

CVM
CVM

NDFS
Figure 4 Elastic Deduplication Engine

Nutanix Shadow Clones delivers distributed localized caching of virtual disks

performance in multi-reader scenarios, such as desktop virtualization using VMware
Horizon 6 using Linked Clones. With Shadow Clones, the CVM actively monitors
virtual disk access trends. If there are requests originating from more than two
remote CVMs, as well as the local CVM, and all of the requests are read I/O and the
virtual disk will be marked as immutable. Once the disk has been marked immutable,
the virtual disk is then cached locally by each CVM, so read operations are now
satisfied locally by local storage.

VMware Horizon 6.0 with View | 8

Figure 5 Nutanix Shadow Clone Functionality

VMware Horizon 6.0 with View | 9

Better together: Nutanix and VMware

Benefits of the combined VMware and Nutanix solution include:

Sim ple, out-of-the-box deploym ent: Ready to deploy virtual desktops in under
60 minutes, managed from the virtualization console.

1
0

Linear scale out: Scale users seamlessly and modularly with no performance
degradation. Data locality allows local caching of data to remain close to the
workload and reduces congestion on the network. A control plane that spans all of
the nodes ensures that resources are never stranded like flash and managed
ineffectively.
Better than PC perform ance: NOS features, including inline-deduplication,
eliminate IOPS resulting in fast application response and boot/login experience.
Base images can be fingerprinted enabling the benefits of in-line deduplication with
no overhead. The Nutanix platform provides up to 130,000 plus random read IOPS
and over 78,000 random write IOPS up in a compact 2U 4-node cluster.
Lower costs: Lower infrastructure CAPEX than a PC and lower ongoing operating
costs due to ease of use and small footprint.
Elim inate project risk: Start small and expand as warranted always utilizing the
latest advances in CPU, memory, and flash.
Business continuity: Built-in native replication and disaster recovery (DR)
features enable highly available desktops to be deployed in mission-critical
environments. Block awareness allows larger clusters to lose up to 4-nodes without
using any additional capacity.
Enterprise-grade m anagem ent: Nutanix Prism delivers a simplified and intuitive
consumer-grade approach to managing large clusters, including a converged
management tool that serves as a single pane for servers and storage, alert
notifications, and provides the IPv6 bonjour mechanism to auto-detect new nodes in
the cluster. It provides the ability to spend more time enhancing your environment,
not maintaining it.
Prism Central allows control over multiple clusters, enabling true multi-tenancy and
desktop as service. Prism Central allows the VMware Horizon Cloud Pod
Architecture to be managed with ease and provides the ability to have physical
separation and control over clusters in local or remote datacenters. Businesses can
decide on a deployment plan that works best for them and their users.
VM w are Integration: Support for View Composer Array Integration (VCAI), and
vStorage APIs for Array Integration (VAAI). Due to Nutanixs ability to cache to local
RAM, the View Storage Accelerator doesnt need to be used, saving deployment
time and storage. In some use cases, Nutanix can remove the need for additional
components, like View Composer, to due strong product integration.

VMware Horizon 6.0 with View | 10

Application Delivery: Nutanix provides space savings by using VMCaliber clones

since Horizon View Composer does not support cloning RDS servers. The Controller
Virtual Machine (CVM) running on each Nutanix Node provides rich data services to
prevent the IO blender effect and provide limitless per-VM snapshots independent
of the hypervisor to give peace of mind for mission-critical applications. Nutanix also
provides per-VM monitoring with Cluster Health to prevent problems and ease
troubleshooting.

1
1

Graphics acceleration: Platforms powered with K1 and K2 cards from Nvidia

GRID and Teradici APEX power tough, graphics-intensive desktops.

The benefits of the Nutanix Platform are now exposed to scale out vSphere deployments:

Figure 6 Add one node at a time to meet your needs.

VMware Horizon 6.0 with View | 11

VMware Horizon 6.0 with View

Horizon with View allows organizations to deliver virtualized or remote desktops
and applications through a single platform and support end users with access to all
of their desktops and applications in one place.

1
2

Figure 7 Horizon with View, platform for end user computing.

Desktop virtualization with Horizon View enables organizations to do more with less
and adopt a user-centric, flexible approach to computing. By decoupling
applications, data, and operating systems from the endpointand by moving these
components into the datacenter where they can be centrally managed in your
clouddesktop and application virtualization offers IT a more streamlined, secure
way to manage users and provide agile, on-demand desktop services.

VMware Horizon 6.0 with View | 12

4. Solution Design
With the Horizon View on Nutanix solution you have the flexibility to start small with
a single block and scale up incrementally a node, a block, or multiple blocks at a
time. This provides the best of both worldsthe ability to start small and grow to
massive scale without any impact on performance.
The following section covers the design decisions and rationale for the Horizon View
deployments on the Nutanix Virtual Computing Platform. View Composer linked
clones where testing using

1
3

Table 1: Platform Design Decisions

Item

Detail

Rationale

Minimum Size

3 x Nutanix nodes (3 vSphere hosts)

Minimum size requirement

Scale Approach

Incremental modular scale

Allow for growth from PoC (hundreds

of desktops) to massive scale
(thousands of desktops)

Scale Unit

Node(s), Block(s), or Pod(s)

Granularly scale to precisely meet

capacity demands

General

Scale in node increments

Blocks consisting of 2,000 user
Pods consisting of up to 10,000 users

Infrastructure Services

Small deployments: Shared cluster

Large deployments: Dedicated
cluster (Node A from 3 blocks or a
1350)

Dedicated infrastructure cluster for

larger deployments (best practice)

vSphere
Cluster Size

Up to 12-24 vSphere hosts

(Minimum of 4 hosts)

Isolated fault domains

Clusters per vCenter

Up to 2x24 or 4x12 host clusters

Task parallelization

Datastore(s)

1 x Nutanix datastore per pod

(Horizon with View, SQL Server, VM
clones, etc.)

Nutanix handles I/O

distribution/localization

VMware best practice

n-Controller model

(Max 2058 machines per container)

Nutanix
Cluster Size

Up to 24-48-nodes

Isolated fault domains

Storage Pool(s)

1 x Storage Pool per cluster

Standard practice
ILM handles tiering

VMware Horizon 6.0 with View | 13

Container(s)

1 x Container for VMs

Standard practice

Features/
Enhancements

(Max 2058 machines per container)

Increase CVM Memory to 24-32GB+
(CVM was set to 32GB for the RA)

High Availability limit

Best practice

1
4

Table 2: VMware Design Decisions

Item

Detail

Rationale

Min: 2 (n+1)
Scale: 1 per additional pod

HA for Connection Brokers

Users per Broker

Up to 2,000 users

Horizon View best practice

Load Balancing

F5 or Load Balancer

Ensures availability of controllers

Balances load between brokers

vCPU: 4
Memory: 10 GB
Disk: 60GB vDisk

Standard sizing practice

Up to 5 + 2 spare

Based upon sizing considerations

F5 or other Load Balancer

Ensures availability and balances load

between controllers.

Horizon View
Infrastructure
Connection Brokers(s)

Connection
Brokers
Virtual Hardware
Specs

Connection Broker(s)
per Pod
Load Balancing

vCenter
vCenter Appliance
1appliance per 2000 VMs
Task Parallelization
5.5.0.10100 Build
Installed separately from vCenter
1750781
Virtual Hardware
vCPU: 8
Resources for fast provisioning
Specs
RAM: 12 GB
Note: Due to the smaller environment one vCenter was shared
between the management and desktop\RDS infrastructure.
Larger environments should have separate vCenter systems for
management and desktop clusters.
View Composer Services
View Composer

1 per vCenter
Installed separately from vCenter

Best Practice

Virtual Hardware
vCPU: 2
Specs
RAML 4 GB
View Security Servers

Best Practice

View Security
Servers(s)
Virtual Hardware
Specs
Load Balancing

Min: 2 (n+1)

HA for security servers

vCPU: 4
RAM: 10 GB
F5 or Load Balancer

Resources for fast provisioning

Ensures availability of storefront
servers
Balances load between storefront
servers & pods

VMware Horizon 6.0 with View | 14

Table 3: Infrastructure Design Decisions

Item

Detail

Rationale

Global Catalog/DNS
Server(s)
DHCP

Min: 2 (n+1) per site

HA for GC/DNS
Microsoft Best Practice

DHCP Server(s)

Min: 2 (n+1) per site

HA for DHCP Servers

Load Balancing

DHCP Server Failover Relationship

Ensures availability of DHCP Servers

Balances load between DHCP Servers
in operation

DFS Server(s)

Min: 2 (n+1) per site

HA for DFS Servers

Load Balancing

Lowest Cost

Ensures availability of DFS

Balances load between DFS Servers

SQL Server(s)

Min: 2 (n+1) per site

Scale: 2 per additional pod

HA for SQL Servers

Data Protection

SQL AlwaysOn Availability Group

Ensures availability of SQL Servers

Active Directory

1
5

File Services

SQL Server

Table 4: Network Design Decisions

Item

Detail

Rationale

vSwitchNutanix

Use: vSphere to CVM local

communication
Uplink(s): N/A

Nutanix Default

vSwitch0 \ vDS

Use: All external VM

communication
Uplink(s): vmnic2,vmnic3

Nutanix Default

NIC(s): 2 x 10Gb
Teaming mode:
standard vSwitch: Port ID
Distrusted vSwitch: Load based
Teaming

Utilize both 10Gb adapters

active/active

ID: Varies
Mask: /24
Components:
vSphere Hosts
Nutanix CVMs
vCenter
SQL Servers
AD/ DHCP/DFS Servers
View Connection Servers

Dedicated infrastructure VLAN

Best Practice

Virtual Switches

NIC Teaming
NetAdapterTeam

VLANs
Management VLAN

VMware Horizon 6.0 with View | 15

vMotion VLAN

ID: Varies
Mask: /24
Components:
vSphere Hosts

vSphere Best Practice

Front-end VLAN(s)

ID: DMZ (for external)

Mask: Varies
Components:
Storefront

Network segmentation for front-end

or external services

1
6

Desktop & RDS Sizing

Nutanix can host both virtual desktops and remote application services. Densities
will vary based upon specific images and workload. For VDI testing, we used a
medium workload using LoginVSI. To test Horizon View RDS functionality, shared
hosted desktops were created to test loads using a LoginVSI 4.1 light workload. The
task worker represents a comparable workload when determining Remote App
densities due to a limited number of applications launched per user and no heavy
video being played.
The following are examples of some typical scenarios for desktop deployment and
utilization.
Table 5: Desktop Scenario Definition

Scenario

Definition

Task Workers

Task workers and administrative workers perform repetitive

tasks within a small set of applications, usually at a stationary
computer. The applications are usually not as CPU and
memory-intensive as the applications used by knowledge
workers. Task workers who work specific shifts might all log in
to their virtual desktops at the same time. Task workers include
call center analysts, retail employees, and warehouse workers.

Knowledge
Workers

Knowledge workers daily tasks include accessing the Internet,

using email, and creating complex documents, presentations,
and spreadsheets. Knowledge workers include accountants,
sales managers, and marketing research analysts.

Power Users

Power users include application developers and people who

use graphics-intensive applications.

The following table contains initial recommendations for desktop sizing for a
Windows 7 desktop. Note: These are recommendations for sizing and should be
modified after a current state analysis.

VMware Horizon 6.0 with View | 16

Table 6: Desktop Scenario Sizing

Scenario
Task Workers
Knowledge Workers
Power Users

vCPU
1
1-2
2

Memory
1GB
2GB
4GB

Disks
30GB (OS)
30GB (OS)
30GB+ (OS)

1
7

Desktop and RDS Optimizations

Following are some high-level desktop optimizations we followed for this design:

Size desktops appropriately for each particular use case.

Use a mix of applications installed in gold images and application

virtualization, depending on the scenario.

Disable unnecessary OS services and applications.

Redirect home directories or use a profile management tool for user profiles
and documents.

Set a page file size.

Table 7: Virtual Desktop - Node Sizing Estimates

Node Type
Virtual Desktop

Task
120

Workload/User Density
Medium
Heavy
100
60

Table 8 RDS - Node Sizing Estimates

Node Type
RDS(shared hosted
desktops)

Light
200

Workload/User Density
Medium
Heavy
170
90

The following figure shows an example of a RDS node providing hosted shared
desktops:

VMware Horizon 6.0 with View | 17

1
8

Figure 8 RDS node providing hosted shared desktops.

Horizon View Composer

Horizon View Composer utilizes a standardized model for hosted virtual desktop
creation. Leveraging a base, or Golden Image, View Composer will create clone
VMs which consist of a delta and identity disk, which links back to the base VMs
disks.
The following figure shows the main architectural components of a Horizon View
deployment on Nutanix and the communication path between services.

Figure 9 Connectivity with View Composer

VMware Horizon 6.0 with View | 18

View Composer with Shadow Clones and VCAI

The next figure describes the high-level IO path for a View Composer based desktop
on Nutanix. As shown, all IO operations are handled by NDFS and occur on the local
node to provide the highest possible IO performance. Read requests will occur
locally for desktops hosted on the same vSphere node as the Replica VM and over
10GbE for desktops hosted on other nodes. The Replica VM is created based off the
selected golden image by ensuring a locked-in state for the duration of its use. It is
possible for the Replica VM to become the bottleneck for performance.

1
9

Figure 10 View Composer IO Overview

Shadow Clones
The following figure describes the detailed IO path for a Horizon with View based
desktop on Nutanix. All write IOs will occur locally on the local nodes SSD tier to
provide the highest possible performance. Read requests for the Replica VM will
occur locally for all desktops when the NDFS Shadow Clone feature is enabled, as
this enables distributed caching of the Replica VM. These reads are served from the
high performance read cache (if cached) or the SSD tier. Each node will also cache
frequently accessed data in the read cache for any local data (delta disks, personal
vDisks (if used)). Nutanix ILM will continue to constantly monitor data and the IO
patterns to choose the appropriate tier placement. This helps to eliminate any
performance bottlenecks.

VMware Horizon 6.0 with View | 19

2
0

Figure 11 Shadow Clones IO Detail

Nutanix Support for View Composer API for Array Integration

Full clones are made when View Composer calls the NFS VAAI (vSphere API for
Array Integration). The full clones are made by taking a copy of the golden image,
known as the replica. Full clones have their own metadata for full read and write
operations. VMs created with VCAI can use the local storage controller and
eliminate any bottlenecks that might happen by having to rely on the replica for
reads. The rest is the same as before, reads are served from the high performance
read cache (if cached) or the SSD tier. Each node will also cache frequently
accessed data in the read cache for any local data (delta disks, personal vDisks (if
used)). Nutanix ILM will continue to constantly monitor data and the IO patterns to
choose the appropriate tier placement. This helps to eliminate any performance
bottlenecks.

Figure 12 View Composer with VCAI IO Detail

VMware Horizon 6.0 with View | 20

Nutanix Web-scale Converged Infrastructure

The Nutanix web-scale converged infrastructure provides an ideal combination of
both high-performance compute with localized storage to meet any demand. True
to this capability, this reference architecture contains zero reconfiguration of or
customization to the Nutanix product to optimize for this use case.
The next figure shows a high-level example of the relationship between a Nutanix
block, node, storage pool, and container:

2
1

Figure 13 Nutanix Component Architecture

The following table shows the Nutanix storage pool and container configuration.

Table 9: Nutanix Storage Configuration

Name
SP01
CTR-RF2-VDI-01
CTR-RF2-RDS-01

Role
Main storage pool for all data
Container for all Desktops
Container for all Servers

Details
All Disks
vSphere Datastore
vSphere Datastore

Network
Designed for true linear scaling, Nutanix leverages a Leaf-Spine network
architecture. A Leaf-Spine architecture consists of two network tiers: an L2 Leaf
and an L3 Spine based on 40GbE and non-blocking switches. This architecture
maintains consistent performance without any throughput reduction due to a static
maximum of three hops from any node in the network.
The following figure shows a design of a scale-out Leaf-Spine network architecture
that provides 20Gb active throughput from each node to its Leaf and scalable 80Gb

VMware Horizon 6.0 with View | 21

active throughput from each Leaf to Spine switch providing scale from 1 Nutanix
block to thousands without any impact to available bandwidth:

2
2

Figure 14 Leaf Spine Network Architecture

Logical Network Design

Each vSphere host has two default switches for internal and external
communication. The standard vSwitch or vDS switch is utilized for external node
communication and VM traffic and has 10GbE uplinks in a team. The vSwitch
Nutanix is utilized for NFS I/O between the vSphere host and the Nutanix CVM.
The following figure shows a logical network representation of the network
segments used in the solution and corresponding components attached.

VMware Horizon 6.0 with View | 22

2
3

Figure 15 Logical Network Connectivity

VMware Horizon 6.0 with View | 23

5. Validation and Benchmarking

The solution and testing provided in this document were completed with VMware
Horizon 6 with View deployed on vSphere 5.5 U1 on the Nutanix Virtual Computing
Platform.
The Login VSI 4.1 medium workload was leveraged to detail the desktop
performance and light workload was leveraged for RDS.

2
4

Nutanix Configuration
A Nutanix NX-3460 was used to host all infrastructure and Horizon services, as well
as the Login VSI test harness. Active Directory services ran inside of the
infrastructure cluster as well DHCP and SQL.
Two Nutanix NX-3460 were utilized as the target environment and provided all
desktop and RDS hosting. Tests were ran with:

4-node VDI & RDS

6-node VDI

8-node VDI

Testing shows that Nutanix can handle Tier 1 workloads and scale in a linear fashion.
Both Nutanix blocks were connected to a top-of-rack switch via 10GbE.

Figure 16 Test Environment Overview

VMware Horizon 6.0 with View | 24

Test Environment Configuration

Assumptions:
o

Medium workload for virtual desktops, light workload for RDS

VDI using View Composer with VCAI enabled

RDS using Nutanix Quick Clones for quick provisioning.

2
5

Hardware:
o

Storage/Compute:, 2 Nutanix NX-3460, NOS 4.0.1

Network: Arista 7050Q(L3 Spine)/7050S(L2 Leaf) Series Switches

Login VSI:
o

Login VS 4.1.0.757 Professional

Horizon View Configuration:

Table 10: Horizon View Configuration

VM

Q ty

vCPU

M em ory

Disks

Connection Brokers

10 GB

1 x 60GB (OS)

View Composer

4 GB

1 x 60GB (OS)

SQL

8 GB

3 X 60 GB (OS,
DATA, Logs)

vCenter Appliance

10 GB

1 X 80 GB, 1 X 100
GB

:
Table 11: Horizon View Test Image Configuration

Attribute

VDI -Window 7 SP1 64 bit

RDS -Window 2012 R2

Hardware

10 GB

1 x 60GB (OS)

CPU

2 vCPUs

N/A

Memory

8 GB

3 X 60 GB (OS, DATA,

VMware Horizon 6.0 with View | 25

Logs)
Memory reserved

10 GB

1 X 80 GB, 1 X 100 GB

Video RAM

128 MB

128MB

3D Graphics

Off

Off

NICs

Virtual network adapter

VMXNet3 Adapter

VMXNet3 Adapter

Virtual SCSI controller 0

Paravirtual

Paravirtual

Virtual disk VMDK1

50 GB

100 GB

Virtual floppy drive

Removed

Removed

Virtual CD/DVD drive 1

Removed

Removed

Applications

Adobe Acrobat 11

Adobe Acrobat 11

Adobe Flash Player 11

Adobe Flash Player 11

Doro PDF 1.82

Doro PDF 1.82

FreeMind

FreeMind

Internet Explorer 11

Internet Explorer 11

MS Office 2010

MS Office 2010

VMware Tools

9.4.5.173405

9.4.5.173405

VMware View Agent

6.0

6.0

2
6

Login VSI Benchmark

Login Virtual Session Indexer (Login VSI) is the de facto industry standard
benchmarking tool to test the performance and scalability of centralized Windows
desktop environments, including Server Based Computing (SBC) and Virtual
Desktop Infrastructures (VDI).
Login VSI is 100% vendor independent and is used to test virtual desktop
environments including VMware Horizon View, or any other Windows-based SBC or
VDI solution.

VMware Horizon 6.0 with View | 26

Login VSI is used for testing and benchmarking by all major hardware and software
vendors, and is recommended by both leading IT analysts and the technical
community. Login VSI is vendor-independent and works with standardized user
workloads, therefore conclusions that are based on Login VSI test data are
objective, verifiable, and replicable.
For more information about Login VSI visit http://www.loginvsi.com/

2
7

For more information about Login VSI test workflows visit

http://www.loginvsi.com/pdf/documentation/Login-VSI-40-Workloads.pdf

Interpreting the Login VSI Results

Login VSI is a test benchmark used to simulate real world user workload on a
desktop. These values are the full time it takes for an application or task to complete
(for example, launch Word) and are not in addition to traditional desktop response
times. These do not refer to the round trip time (RTT) for network IO, rather the
total time to perform an action on the desktop.
During the test, all VMs were powered on and the workload used a launch window
of 2,880 second for all tests.
Evaluation was quantified using the following metrics:
o

Minimum Response: Minimum response indicates the minimum response time

for all the measurements taken when the indicated number of sessions on the
X-axis were active.

Average Response: Average response indicates the average response time

for all the measurements taken when the indicated number of sessions on the
X-axis were active.

Maximum Response: Maximum response indicates the maximum response

time for all the measurements taken when the indicated number of sessions
on the X-axis were active.

VSImax v4 detailed: The individual measurements taken during a test in a

combined graph. This graph shows the minimum, average, and maximum
response times for each individual measurement. There is also a Total metric
that combines all of the metric into a single number. The minimum, average,
and maximum for this combined value is shown as well.

VSI Index Average: Indicates the average value as calculated by VSI. The VSI
Index Average differs from Average Response on the fact that Average
Response is the pure average. VSI Index Average applies certain statistical
rules to the average to avoid spikes from influencing the average too much.

VSImax v4: Shows the amount of sessions can be active on a system before
the system is saturated. The blue X shows the point where VSImax was
reached. This number provides an indication of the scalability of the
environment (higher is better).

VMware Horizon 6.0 with View | 27

VSIbase(line): shows the VSI index average for the environment when there is
no to little load on the environment. This number is used as an indication of
the base performance of the environment (lower is better). This number in
combination with the VSImax number will tell you:
o

How well an environment performs (VSIbase)

And how long the environment can maintain that performance, how
scalable the VSIbase performance is (VSImax).

2
8

LogonTimer: An indication of the time it takes for a session to logon. The

graph shows the trend of logon times during the test. The logon time is
specified in seconds. Please note that this is an indication of the logon time.
VSI measures the time from the logon scripts running, shortly after group
policy has been processed but before the shell has loaded (Windows
Explorer), and the windows shell being loaded.

Based on user experience and industry standards, Nutanix recommends that the
values be kept below the following values (LoginVSI 4.x):

M etric
Minimum Response

Table 12: Login VSI Metric Values

Value(m s)
Rationale
<1,000
Acceptable ideal response time

Average Response
Maximum Response
VSI Baseline
VSI Index Average

<4,000
<8,000
<5,000
<4,000

Acceptable
Acceptable
Acceptable
Acceptable

average response time

peak response time
ideal response time
average response time

Login VSI Graphs

The Login VSI graphs show the values defined obtained from the launching of each
desktop session. The following figure shows an example graph showing the test
data. The y-axis is the response time in milliseconds and the x-axis is the number of
active sessions.

VMware Horizon 6.0 with View | 28

2
9

Figure 17 Sample LoginVSI graph

VMware Horizon 6.0 with View | 29

6. Validation Results
All tests had 111 users launched per node with the objective of getting 110 or more
users to run as the Nutanix cluster scaled from 4,6,8-nodes proving the linear
scalability.

View Composer with VCAI for a 4-node NX-3060 (NX-3460)

3
0

The following are the test results for a 4-node NX-3460 with View Composer Array
Integration enabled with LoginVSI medium user profile supported by Window 7 SP1
Login VSI Medium Results
During the testing, VSImax was not reached with a baseline of 1346 and average
VSImax of 3376ms. The VSImax threshold was 3946. 441 sessions ran successfully.

Figure 18 VSImax for a 441 Horizon View users on 4-nodes.

VMware Horizon 6.0 with View | 30

CVM CPU Metrics

At the peak of the test execution, CVM CPU utilization for the vSphere hosts
averaged at 53%. During the boot phase of all 36 Servers CVM usage did reach 72%

3
1

Figure 19 CPU Utilization for all CVMs during the testing for 4-nodes.

Nutanix Storage Metrics

The 4-node tests showed IO footprints on the Nutanix platform with a peak of 3,65
aggregate IOPS during the test runs. IO latencies averaged < 2.4 ms and peaked at
3.1 ms during the test.

VMware Horizon 6.0 with View | 31

Figure 20 4 Node - Total Cluster Read &Write IOPS during testing

3
2

Figure 21 4-node IO latency during testing with a peak of 3.1ms

Overall Cluster CPU and Memory Consumption

The overall CPU did reach 93.32% during the peak of the run and overall memory ~
80% utilized across the 4-nodes.

Figure 22 Overall Cluster CPU and Memory Consumption for 4-nodes

View Composer with VCAI for a 4-node NX-3060 (NX-3460+NX-3260)

The following are the test results for a 6-node configuration using NX-3060 nodes
with View Composer Array Integration enabled with LoginVSI medium user profile
supported by Window 7 SP1.

VMware Horizon 6.0 with View | 32

Login VSI Medium Results

During the testing, VSImax was reached with a baseline of 1264 and average VSImax
of 2708ms. The VSImax threshold was 3864. VSImax was 664 users.

3
3

Figure 23 VSImax for a 664 Horizon View users on 6 nodes.

CVM CPU Metrics

At the peak of the test execution, CVM CPU utilization for the vSphere hosts
averaged at 65%. During the peak of testing with 6 nodes, CPU utilization peaked at
99.68%

VMware Horizon 6.0 with View | 33

Figure 24 CPU Utilization for all CVMs during the testing for 6 nodes with 664 users.

3
4

Nutanix Storage Metrics

The 6-node tests showed IO footprints on the Nutanix platform with a peak of 5,740
aggregate IOPS during the test runs. IO latencies averaged < 1.7 ms and peaked at
2.21 ms during the test.

Figure 25 Aggregate cluster IOPS for 6 nodes

View Composer with VCAI with 8-nodes NX-3060 (2x NX-3460)

The following are the test results for an 8-node configuration using NX-3060-nodes
with View Composer Array Integration enabled with LoginVSI medium user profile
supported by Window 7 SP1
Login VSI Medium Results
During the testing VSImax was not reached with a baseline of 1293 and average
VSImax of 3690ms. The VSImax threshold was 3893. The 8-node configuration
using NX-3060 was able to support 886 user sessions.

VMware Horizon 6.0 with View | 34

3
5

Figure 26 VSImax for 886 Horizon View users on 8-nodes.

CVM CPU Metrics

At the peak of the test execution, CVM CPU utilization for the vSphere hosts
averaged at 49%. During the peak of testing with 8-nodes CPU utilization peaked at
97.95%.

Figure 27 CPU Utilization for all CVMs during the testing for 6-nodes with 664 users.

VMware Horizon 6.0 with View | 35

Nutanix Storage Metrics

The 8-node tests showed IO footprints on the Nutanix platform with a peak of 8,150
aggregate IOPS during the test runs. IO latencies averaged < 1.9 ms and peaked at
2.7 ms during the test.

3
6

Figure 28 8 Node - Total Cluster Read & Write IOPS during testing

Figure 29 6-node - IO latency during testing with a peak of 2.7ms

Overall Cluster CPU and Memory Consumption

The overall CPU did reach 94.68% during the peak of the run and overall memory ~
84% utilized across the 8-nodes.

VMware Horizon 6.0 with View | 36

Figure 30 Overall Cluster CPU and Memory Consumption for 8 Nodes

3
7

Linear Scale
During testing the 4-, 6- and 8-nodes test runs for Horizon View the number of users
per node and VSImax stayed consistent. Small variance can be contributed to
background cluster tasks that are allowing running to ensure high available for the
cluster like self-healing.

Figure 31 Linear Scale as nodes are added to the cluster.

Boot Storm
Boot storms can be avoided or planned for but bad things happen like power
outages, maintenance windows and if youre dealing with events like shift change,
like in Health Care, you need to have ability to boot your desktops quickly. The
clock started at 2:25:34 in vCenter and the watch stopped when all the agents
reported back in the Horizon View Connection broker at 2:31:32. The system
required 6 minutes to boot 888 desktops running on 8 Nutanix NX-3460 nodes.

VMware Horizon 6.0 with View | 37

Figure 32 Cluster CPU did hit high briefly with the current vCenter settings but thats to be expected.

3
8

Figure 33 Minimal latency when booting all of the desktops

Figure 34 Over 50,000 IOPS to boot the desktops. Most of the IOPS coming from local cache.

High Availability and Continuity

One of the added benefits of scale-storage is the addition of multiple storage
controllers. When you have more than 2 storage controllers and you lose one due to
failure or maintenance like a rolling upgrade, you can do so with minimal impact.
Below are the results of 8-node cluster with 700 desktops running a LoginVSI
medium workload. One of the 8 storage controllers is shutdown to see the impact
on the cluster. No desktops were rebooted or shutdown. IOPS dropped 2,000 to
1,496 and latency had a brief spike from 4ms to 22.37 ms.

VMware Horizon 6.0 with View | 38

3
9

Figure 35 Impact of shutting off a Nutanix Storage Controller

Remote Desktop Services on NX-3460 (4-nodes)

The following are the test results for a 4-node configuration using NX-3060 nodes
with View Composer Array Integration enabled with LoginVSI light user profile
supported by Windows 2012 R2. The task worker profile best represents Horizon
with Views delivery of remote applications to users. Each Nutanix node had 8x
Windows 2012 R2 VMs with 5 vCPU and 24 GB of RAM each.
Login VSI Light Workload Results
During the testing, VSImax was reached with a baseline of 1084 and average VSImax
of 3606. The VSImax threshold was 3685. The 4-node Nutanix NX-3460 was able to
support 799 user sessions.

Figure 36 VSImax for 799 Horizon View RDS users on 4-nodes.

VMware Horizon 6.0 with View | 39

CVM CPU Metrics

At the peak of the test execution, CVM CPU utilization for the vSphere hosts
averaged at 41%. During the peak of testing with 4-nodes CPU utilization peaked at
49%.

4
0

Figure 37 CPU Utilization for all CVMs during the testing for 6 nodes with 664 users.

Nutanix Storage Metrics

The 4-node tests showed light IO footprints on the Nutanix platform with a peak of
1,313 aggregate IOPS during the test runs. IO latencies averaged < 4.8 ms and
peaked at 5.62 ms during the test.

Figure 38 4 Node RDS - Total Cluster Read & Write IOPS during testing

VMware Horizon 6.0 with View | 40

4
1

Figure 39 4-node RDS - IO latency during testing with a peak of 5.79 ms

VMCaliber Clone - Time to Provision

Table 13: Time to Provision

Time to take a full clone using

Nutanix VMCaliber
Windows Server 2012 R2 120 GB
VMDK

Start Time
11:33:16

End Time
11:33:24

Total Time
8 sec

With Integration of VAAI (vStorage APIs for Array Integration) support, Nutanix can
clone large servers like Windows Server 2012 R2 being used for RDS in seconds. This
is very important as Horizon View relies on host/virtualization for deployment of
RDS host. The ability to use Nutanix VMCaliber clones also saves flash for
performance instead of being used for copy operations.

VMware Horizon 6.0 with View | 41

7. Solution Application
This section applies this pod-based reference architecture to real-world scenarios
and outlines the sizing metrics and components. The applications below assume a
standard medium user workload, however will vary based upon utilization and
workload.

4
2

NO TE: Detailed hardware configuration and product models can be found in the
appendix. Any starting size of 3 or more nodes can form the base of a Nutanix
cluster.

Scenario: 12 Nodes
Allocating 8-nodes for VDI and 4-nodes for RDS.
Table 14: Detailed Component Breakdown 12 Nodes

Item
Components
# of Nutanix nodes
# of Nutanix blocks
# of RU (Nutanix)
# of 10GbE Ports
# of 100/1000 Ports (IPMI)
# of L2 Leaf Switches
# of L3 Spine Switches

Value
12
3
6
24
12
2
1

Item
Infrastructure
# of vCenter Servers
# of vSphere Hosts
# of vSphere Clusters
# of Datastore(s)
# of RDS Servers
# of RDS Users
# of Virtual Desktops

Value
1
12
2
1
Up to 32
800
880

VMware Horizon 6.0 with View | 42

4
3

Figure 40 Rack Layout 12 Nodes

Availability Domains
Availability Domains is a key construct for distributed systems to abide by for
determining component and data placement. When you have 3 or more uniform
blocks each with a minimum of 2 nodes you can lose a block and the cluster will
keep running. No additional capacity is used to achieve this redundancy just
intelligent data placement.
With the following configuration, a whole block can be down and the Nutanix
Cluster will still run. This allows for more than one node to be down for maintenance
at time, which leads to better OPEX during maintenance windows and overall higher
availability.

VMware Horizon 6.0 with View | 43

4
4

Figure 41 12-node availability domain with VDI & RDS

Scenario: 24 Nodes
This configuration comprises of 16 NX-3060 nodes for VDI and 8 NX-3060 nodes for
RDS.

Table 15: Detailed Component Breakdown 24 Nodes

Item
Components
# of Nutanix blocks
# of Nutanix nodes
# of RU (Nutanix)
# of 10GbE Ports
# of 100/1000 Ports (IPMI)
# of L2 Leaf Switches
# of L3 Spine Switches

Value
6
24
12
48
24
2
1

Item
Infrastructure
# of vCenter Servers
# of vSphere Hosts
# of vSphere Clusters
# of Datastore(s)
# of RDS Servers
# of RDS Users
# of VDI

Value
2
24
1-2
1
Up to 128
Up to
1,600
1,600

VMware Horizon 6.0 with View | 44

4
5

Figure 42 Rack Layout 24 Nodes

VMware Horizon 6.0 with View | 45

Availability Domains
With the following configuration a whole block can be down and the Nutanix Cluster
will still run. This allows for more than one node to be down for maintenance at time,
which leads to better OPEX during maintenance windows and overall higher
availability.

4
6

Figure 43 24-node availability domain with VDI & RDS

VMware Horizon 6.0 with View | 46

8. Conclusion
The VMware Horizon with View and Nutanix solution provides a single high-density
platform for desktop and application delivery. This modular linearly scaling
approach enables these deployments to easily grow. Localized and distributed
Shadow Clone cache, VAAI support allows RDS and Full Clone desktops to be
quickly deployed without wasting high-performance flash. Robust self-healing and
multi-storage controllers deliver high availability in the face of failure or rolling
upgrades.

4
7

Horizon user density will be primarily driven by the available host CPU resources
and not due to any IO or resource bottleneck for both virtual desktops and RDS
deployments on Nutanix. Login VSI test results showed densities of over 110 users
per Nutanix node for VDI and almost 200 users per node for RDS. From starting at
4-nodes and then scaling from 6- to 8-nodes, testing shows that Nutanix can offer a
pay as you grow model like public cloud providers, but in the comfort and security
of your own premises.
By having Nutanix Clusters with both RDS and virtual desktops, you can achieve
greater resiliency by allowing enough nodes to meet the requirements for
Availability Domains. The ability to lose up to 4 nodes without downtime comes
without sacrificing storage capacity and allows IT operations teams to have smaller
maintenance windows.
Infrastructure can be added on your terms and on your schedule without sacrificing
performance or overspending upfront.

VMware Horizon 6.0 with View | 47

9. Appendix: Configuration
Hardware
o

2 * Nutanix NX-3460
o

4
8

Storage / Compute
Per node specs (4-nodes per 2U block):

CPU: 2x Intel Xeon E5-2670

Memory: 256 GB Memory

Network

Arista 7050Q - L3 Spine

Arista 7050S - L2 Leaf

Software
o

Nutanix

NOS 4.0.1

CVM 8vCPUs, 32 GB of RAM

Horizon View

6.0

Virtual Desktop
o

RDS Server

Windows 7 SP1

Microsoft Windows Server 2012 R2

Infrastructure

vSphere 5.5 U1 Build 1881737

VM
o

RDS

CPU: 5 vCPU

Memory: 24 GB (static)

Storage:

1 x 120 GB OS Disk on CTR-RF2-RDS-01 NDFS backed NFS datastore

VMware Horizon 6.0 with View | 48

10. References
Table of Figures
Figure 1 Web-scale properties of Nutanix ....................................................................................... 6

4
9

Figure 2 Nutanix Node Architecture .................................................................................................. 7

Figure 3 Nutanix Architecture .............................................................................................................. 7
Figure 4 Elastic Deduplication Engine .............................................................................................. 8
Figure 5 Nutanix Shadow Clone Functionality .............................................................................. 9
Figure 6 Add one node at a time to meet your needs. ............................................................. 11
Figure 7 Horizon with View, platform for end user computing............................................. 12
Figure 8 RDS node providing hosted shared desktops. .......................................................... 18
Figure 9 Connectivity with View Composer ................................................................................. 18
Figure 10 View Composer IO Overview .......................................................................................... 19
Figure 11 Shadow Clones IO Detail ................................................................................................. 20
Figure 12 View Composer with VCAI IO Detail .......................................................................... 20
Figure 13 Nutanix Component Architecture ................................................................................. 21
Figure 14 Leaf Spine Network Architecture ................................................................................. 22
Figure 15 Logical Network Connectivity ........................................................................................ 23
Figure 16 Test Environment Overview ........................................................................................... 24
Figure 17 Sample LoginVSI graph .................................................................................................... 29
Figure 18 VSImax for a 441 Horizon View users on 4-nodes. ................................................ 30
Figure 19 CPU Utilization for all CVMs during the testing for 4-nodes. ............................ 31
Figure 20 4 Node - Total Cluster Read &Write IOPS during testing ................................. 32
Figure 21 4-node IO latency during testing with a peak of 3.0 ms ..................................... 32
Figure 22 Overall Cluster CPU and Memory Consumption for 4-nodes ........................... 32

VMware Horizon 6.0 with View | 49

Figure 23 VSImax for a 664 Horizon View users on 6 nodes. ............................................... 33

Figure 24 CPU Utilization for all CVMs during the testing for 6 nodes with 664 users.34
Figure 25 Aggregate cluster IOPS for 6 nodes ........................................................................... 34
Figure 26 VSImax for a 886 Horizon View users on 8-nodes................................................ 35
Figure 27 CPU Utilization for all CVMs during the testing for 6-nodes with 664 users.35

5
0

Figure 28 8 Node - Total Cluster Read & Write IOPS during testing ................................. 36
Figure 29 6-node - IO latency during testing with a peak of 2.9 ms .................................. 36
Figure 30 Overall Cluster CPU and Memory Consumption for 8 Nodes........................... 37
Figure 31 Linear Scale as nodes are added to the cluster. ..................................................... 37
Figure 32 Cluster CPU did hit high briefly with the current vCenter settings but thats
to be expected. ........................................................................................................................................ 38
Figure 33 Minimal latency when booting all of the desktops ................................................ 38
Figure 34 Over 50,000 IOPS to boot the desktops. Most of the IOPS coming from
local cache. ................................................................................................................................................ 38
Figure 35 Impact of shutting off a Nutanix Storage Controller............................................ 39
Figure 36 VSImax for 799 Horizon View RDS users on 4-nodes. ........................................ 39
Figure 37 CPU Utilization for all CVMs during the testing for 6 nodes with 664 users.40
Figure 38 4 Node RDS - Total Cluster Read & Write IOPS during testing ...................... 40
Figure 39 4-node RDS - IO latency during testing with a peak of 5.62 ms ..................... 41
Figure 40 Rack Layout 12 Nodes .................................................................................................... 43
Figure 41 12 node availability domain with VDI & RDS............................................................ 44
Figure 42 Rack Layout 24 Nodes.................................................................................................. 45
Figure 43 24-node availability domain with VDI & RDS ..........................................................46

Table of Tables
Table 1: Platform Design Decisions ................................................................................................... 13

VMware Horizon 6.0 with View | 50

Table 2: VMware Design Decisions ................................................................................................... 14

Table 3: Infrastructure Design Decisions ........................................................................................ 15
Table 4: Network Design Decisions .................................................................................................. 15
Table 5: Desktop Scenario Definition .............................................................................................. 16
Table 6: Desktop Scenario Sizing ...................................................................................................... 17

5
1

Table 7: Virtual Desktop - Node Sizing Estimates ...................................................................... 17

Table 8 RDS - Node Sizing Estimates .............................................................................................. 17
Table 9: Nutanix Storage Configuration ......................................................................................... 21
Table 10: Horizon View Configuration ............................................................................................ 25
Table 11: Horizon View Test Image Configuration ...................................................................... 25
Table 12: Login VSI Metric Values..................................................................................................... 28
Table 13: Time to Provision .................................................................................................................. 41
Table 14: Detailed Component Breakdown 12 Nodes ........................................................... 42
Table 15: Detailed Component Breakdown 24 Nodes ........................................................... 44

VMware Horizon 6.0 with View | 51

11. About the Authors

Dwayne Lessner is a technical marketing engineer on the product marketing team at
Nutanix, Inc. In this role, Dwayne helps design, test, and build solutions on top of the
Nutanix Virtual Computing Platform. Dwayne has worked in healthcare and oil & gas
for over ten years in various roles. A strong background in server and desktop
virtualization has given Dwayne the opportunity to work with many different
applications frameworks and architecture. Dwayne has been a speaker at BriForum
and various VMUG events and conferences.

5
2

Steven Poitras is a solution architect on the technical marketing team at Nutanix,

Inc. In this role, Steven helps design architectures combining applications with the
Nutanix platform, creating solutions helping solve critical business needs and
requirements and disrupting the infrastructure space. Prior to joining Nutanix, he
was one of the key solution architects at the Accenture Technology Labs where he
was focused on the Next Generation Infrastructure (NGI) and Next Generation
Datacenter (NGDC) domains. In these spaces, he has developed methodologies,
reference architectures, and frameworks focusing on the design and transformation
to agile, scalable, and cost-effective infrastructures that can be consumed in a
service-oriented or cloud-like manner.

VMware Horizon 6.0 with View | 52

5
3

VMware Horizon 6.0 with View | 53