Cisco Solution for EMC VSPEX End User Computing for 2000 VMware Horizon View 5.2 Users
Available Languages
Table Of Contents
About Cisco Validated Design (CVD) Program
Cisco Solution for EMC VSPEX End User Computing
Benefits of Cisco Unified Computing System
Benefits of EMC VNX Family of Storage Controller
Benefits of VMware Horizon View 5.2
Cisco Unified Computing System
Cisco Unified Computing System Components
Cisco UCS 2200 Series IO Module
Cisco UCS B200 M3 Blade Server
Cisco UCS VIC1240 Converged Network Adapter
Enhancements in Horizon View 5.2
VMware Horizon View 5.2 Hosted Virtual Desktop Overview
Modular Virtual Desktop Infrastructure Technical Overview
Cisco Data Center Infrastructure for Desktop Virtualization
Understanding Desktop User Groups
Understanding Applications and Data
Project Planning and Solution Sizing Sample Questions
Proof of Concept Pilot Program
The Solution: A Unified, Pre-Tested and Validated Infrastructure
Cisco Networking Infrastructure
Cisco Nexus 5500 Series Feature Highlights
Cisco Nexus 1000V Feature Highlight
Architecture and Design of Horizon View 5.2 on Cisco Unified Computing System and EMC VNX Storage
Hosted Virtual Desktop Design Fundamentals
Choose a User Profile Management System
Accessing USB Devices Connected to the End Point
Designing a VMware Horizon View 5.2 Deployment
Determine Desktop Pools Required
Hosted Virtual Desktop Infrastructure Design
Cisco Unified Computing System Configuration
Base Cisco Unified Computing System Configuration
QoS and CoS in Cisco Unified Computing System
Steps to Enable QOS on Cisco Unified Computing System
Nexus 1000V Configuration in L3 Mode
Configuring Boot from SAN Overview
SAN Configuration on Nexus 5548UP
Configuring Boot from SAN on EMC VNX
SAN Configuration on Cisco UCS Manager
EMC VNX5500 Storage Configuration
EMC Storage Configuration for VMware View
EMC Storage Configuration for VMware ESXi 5.1 Infrastructure Clusters
Example EMC Boot LUN Configuration
Cisco UCS Manager Configuration for VMware ESXi 5.1
Installing and Configuring ESXi 5.1
ESXi 5.1 Cluster Configuration
Installing and configuring VMware View 5.2
Create SQL Databases for View 5.2
Install View Manager and components
View Administrator Configuration
Install SSL Certificate for View Connection and Replica Server
Install View Client on End Points
Configure the View 5.2 Hosts and Storage
Desktop Delivery Base Image Creation and Desktop Deployment
Microsoft Windows 7 Golden Image Creation
Create base Windows SP1 Virtual Machine
Optimization of Base Windows 7 SP1 Virtual Machine
Install View 5.2 Virtual Desktop Agent Software
Perform Additional View 5.2 Configuration
Configure the View Desktop Pools and Options.
Cisco UCS Test Configuration for Single Blade Scalability
Cisco UCS Configuration for Two Chassis—Fourteen Blade Test
Testing Methodology and Success Criteria
User Workload Simulation—LoginVSI From Login Consultants
Cisco UCS Test Configuration for Single-Server Scalability Test Results
Cisco UCS Test Configuration for 2000 Desktop Two-Cluster Scalability Test Results
Scalability Considerations and Guidelines
Cisco UCS System Configuration Considerations
VMware View 5.2 Considerations
EMC VNX Storage Guidelines for Horizon View 5.2 Provisioned Virtual Machines
VMware ESXi 5.1 Guidelines for Virtual Desktop Infrastructure
VMware View Reference Documents
Cisco Solution for EMC VSPEX End User ComputingFor 2000 VMware Horizon View 5.2 UsersLast Updated: November 21, 2013
Building Architectures to Solve Business Problems
About the Authors
Ramesh Guduru, Support Engineer, Cisco SystemsRamesh Guduru is a Virtualization System Engineer at Cisco with SSVPG. Ramesh has over 7 years of experience with VMware view thin client administration, configuration and optimization of virtual desktop environment. Ramesh is skilled on core VMware applications in the virtual environment focusing in system design and implementation of virtualization components. Ramesh holds certification in virtualization, network and Microsoft.
Hardik Patel, Support Engineer, Cisco SystemsHardik Patel is a Virtualization System Engineer at Cisco with SSVPG. Hardik has over 9 years of experience with server virtualization and core application in the virtual environment with area of focus in design and implementation of systems and virtualization, manage and administration, Cisco Unified Computing System, storage and network configurations. Hardik holds Masters degree in Computer Science with various career oriented certification in virtualization, network and Microsoft.
Acknowledgments
For their support and contribution to the design, validation, and creation of this Cisco Validated Design, we would like to thank:
•
Mike Brennan, Manager, Technical Marketing, Cisco Systems, Inc.
•
•
About Cisco Validated Design (CVD) Program
The CVD program consists of systems and solutions designed, tested, and documented to facilitate faster, more reliable, and more predictable customer deployments. For more information visit
http://www.cisco.com/go/designzone.
ALL DESIGNS, SPECIFICATIONS, STATEMENTS, INFORMATION, AND RECOMMENDATIONS (COLLECTIVELY, "DESIGNS") IN THIS MANUAL ARE PRESENTED "AS IS," WITH ALL FAULTS. CISCO AND ITS SUPPLIERS DISCLAIM ALL WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE WARRANTY OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT OR ARISING FROM A COURSE OF DEALING, USAGE, OR TRADE PRACTICE. IN NO EVENT SHALL CISCO OR ITS SUPPLIERS BE LIABLE FOR ANY INDIRECT, SPECIAL, CONSEQUENTIAL, OR INCIDENTAL DAMAGES, INCLUDING, WITHOUT LIMITATION, LOST PROFITS OR LOSS OR DAMAGE TO DATA ARISING OUT OF THE USE OR INABILITY TO USE THE DESIGNS, EVEN IF CISCO OR ITS SUPPLIERS HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
THE DESIGNS ARE SUBJECT TO CHANGE WITHOUT NOTICE. USERS ARE SOLELY RESPONSIBLE FOR THEIR APPLICATION OF THE DESIGNS. THE DESIGNS DO NOT CONSTITUTE THE TECHNICAL OR OTHER PROFESSIONAL ADVICE OF CISCO, ITS SUPPLIERS OR PARTNERS. USERS SHOULD CONSULT THEIR OWN TECHNICAL ADVISORS BEFORE IMPLEMENTING THE DESIGNS. RESULTS MAY VARY DEPENDING ON FACTORS NOT TESTED BY CISCO.
CCDE, CCENT, Cisco Eos, Cisco Lumin, Cisco Nexus, Cisco StadiumVision, Cisco TelePresence, Cisco WebEx, the Cisco logo, DCE, and Welcome to the Human Network are trademarks; Changing the Way We Work, Live, Play, and Learn and Cisco Store are service marks; and Access Registrar, Aironet, AsyncOS, Bringing the Meeting To You, Catalyst, CCDA, CCDP, CCIE, CCIP, CCNA, CCNP, CCSP, CCVP, Cisco, the Cisco Certified Internetwork Expert logo, Cisco IOS, Cisco Press, Cisco Systems, Cisco Systems Capital, the Cisco Systems logo, Cisco Unity, Collaboration Without Limitation, EtherFast, EtherSwitch, Event Center, Fast Step, Follow Me Browsing, FormShare, GigaDrive, HomeLink, Internet Quotient, IOS, iPhone, iQuick Study, IronPort, the IronPort logo, LightStream, Linksys, MediaTone, MeetingPlace, MeetingPlace Chime Sound, MGX, Networkers, Networking Academy, Network Registrar, PCNow, PIX, PowerPanels, ProConnect, ScriptShare, SenderBase, SMARTnet, Spectrum Expert, StackWise, The Fastest Way to Increase Your Internet Quotient, TransPath, WebEx, and the WebEx logo are registered trademarks of Cisco Systems, Inc. and/or its affiliates in the United States and certain other countries.
All other trademarks mentioned in this document or website are the property of their respective owners. The use of the word partner does not imply a partnership relationship between Cisco and any other company. (0809R)
© 2013 Cisco Systems, Inc. All rights reserved
Cisco Solution for EMC VSPEX End User Computing
Overview
Industry trends indicate a vast data center transformation toward shared infrastructures. Enterprise customers are moving away from silos of information and toward shared infrastructures, to virtualized environments, and eventually to the cloud to increase agility and reduce costs.
This Cisco Solution for EMC VSPEX End User Computing reports the results of a study evaluating the scalability of VMware Horizon View 5.2 environment on Cisco UCS B-Series B200 M3 Blade Servers running VMware ESXi 5.1 hypervisor software connected to an EMC VNX 5500 Storage Array. We utilize second and third generation Cisco Unified Computing System hardware and software. We provide best practice recommendations and sizing guidelines for large scale customer deployments of VMware Horizon View 5.2 on the Cisco Unified Computing System™.
Audience
This document describes the architecture and deployment procedures of an infrastructure comprised of Cisco, EMC, and VMware hypervisor and desktop virtualization products. The intended audience of this document includes, but is not limited to, sales engineers, field consultants, professional services, IT managers, partner engineering, and customers who want to deploy the solution described in this document.
Solution Component Benefits
Each of the components of the overall solution materially contributes to the value of functional design contained in this document.
Benefits of Cisco Unified Computing System
Cisco Unified Computing System™ is the first converged data center platform that combines industry-standard, x86-architecture servers with networking and storage access into a single converged system. The system is entirely programmable using unified, model-based management to simplify and speed deployment of enterprise-class applications and services running in bare-metal, virtualized, and cloud computing environments.
Benefits of the Unified Computing System include:
Architectural flexibility
•
•
•
•
Infrastructure Simplicity
•
•
•
•
Business Agility
•
•
•
Benefits of Nexus 5548UP
The Cisco Nexus 5548UP Switch delivers innovative architectural flexibility, infrastructure simplicity, and business agility, with support for networking standards. For traditional, virtualized, unified, and high-performance computing (HPC) environments, it offers a long list of IT and business advantages, including:
Architectural Flexibility
•
•
•
•
•
•
•
Infrastructure Simplicity
•
•
•
•
•
Business Agility
•
•
•
Specifications At-a-Glance
•
•
•
•
Benefits of EMC VNX Family of Storage Controller
The EMC VNX Family delivers industry leading innovation and enterprise capabilities for file, block, and object storage in a scalable, easy-to-use solution. This next-generation storage platform combines powerful and flexible hardware with advanced efficiency, management, and protection software to meet the demanding needs of today's enterprises.
All of this is available in a choice of systems ranging from affordable entry-level solutions to high performance, petabyte-capacity configurations servicing the most demanding application requirements. The VNX family includes the VNXe Series, purpose-built for the IT generalist in smaller environments, and the VNX Series , designed to meet the high-performance, high scalability, requirements of midsize and large enterprises.
VNX Series—Simple, Efficient, Powerful
A robust platform for consolidation of legacy block storage, file-servers, and direct-attached application storage, the VNX series enables organizations to dynamically grow, share, and cost-effectively manage multi-protocol file systems and multi-protocol block storage access. The VNX Operating environment enables Microsoft Windows and Linux/UNIX clients to share files in multi-protocol (NFS and CIFS) environments. At the same time it supports iSCSI, Fiber Channel, and FCoE access for high bandwidth and latency-sensitive block applications. The combination of EMC Atmos Virtual Edition software and VNX storage supports object-based storage and enables customers to manage web applications from EMC Unisphere. The VNX series next generation storage platform is powered by Intel quad-core Xeon 5600 series with a 6 -Gb/s SAS drive back-end and delivers demonstrable performance improvements over the previous generation mid-tier storage:
•
•
•
Benefits of VMware ESXi 5.1
Virtualization is now a critical component to an overall IT strategy, it is important to choose the right vendor. VMware is the leading business virtualization infrastructure provider, offering the most trusted and reliable platform for building private clouds and federating to public clouds.
Find out how VMware delivers on the core requirements for a business virtualization infrastructure solution:
•
•
•
•
•
And best of all, VMware provides:
•
For detailed information about vSphere 5.1, go to:
http://www.vmware.com/files/pdf/products/vsphere/vmware-what-is-new-vsphere51.pdf
Benefits of VMware Horizon View 5.2
Deliver rich, personalized virtual desktops as a managed service from a virtualization platform built to deliver the entire desktop, including the operating system, applications and data. With Horizon View, desktop administrators virtualize the operating system, applications, and user data and deliver modern desktops to end-users. Get centralized automated management of these components for increased control and cost savings. Improve business agility while providing a flexible high performance desktop experience with VMware Horizon View, desktop administrators virtualize the operating system, applications, and user data and deliver modern desktops to end-users, across a variety of network conditions.
Deliver Business Agility
Bring the agility and availability of cloud computing to the desktop and applications with Horizon View. Built on VMware vSphere, Horizon View delivers desktops from a single integrated platform as part of your cloud services.
Dynamically allocate resources to enable highly responsive environment to end users. Scale up and down desktop services on demand to quickly meet changing business needs and proactively protect against planned and unplanned downtime. Run your desktops as business critical services for your workforce.
Easily Control and Manage Desktops
Increase control of desktops, applications and data by delivering and managing them as centralized services.
A single, powerful administrative console provides oversight of desktop services while enabling IT to simply execute previously cumbersome tasks like provisioning, updates and patches. Easily apply policies, quickly enable and disable users all from a centralized console for optimal business response. Free up time from maintenance for technology innovation.
Deliver a Better Desktop Experience
Unlike traditional PCs, View desktops are not tied to the physical computer. Instead, they reside in your cloud and end-users can access their View desktop when needed.
Horizon View with PCoIP delivers the richest, most flexible and adaptive experience for end-users around the world in a variety of network conditions. Business happens everywhere and whether online or offline, desktops or mobile devices, LAN or WAN, Horizon View delivers maximum workplace productivity.
Automate Desktop Operations Management
VMware vCenter Operations Manager for View allows administrators to gain insight into desktop and infrastructure performance, quickly pinpoint and troubleshoot issues. Administrators can optimize resource utilization, and proactively manage the desktop environment through the management dashboards. vCenter Operations Manager for View is an optional add-on for Horizon View customers. You can also leverage PCoIP Extension Services to collect Horizon View statistics into your existing WMI tool.
Built-in Security
Maintain control over data and intellectual property by keeping it secure in the datacenter. Encrypted protocol traffic provides secure end-users access virtual desktops inside or outside of the corporate network. Integration with vShield Endpoint enables offloaded and centralized anti-virus and anti-malware (AV) solutions. This integration helps to eliminate agent sprawl and AV storm issues while minimizing the risk of malware infection and simplifying AV administration. VMware View also supports integration with RSA SecureID for 2-factor authentication requirements.
Summary of Main Findings
The combination of technologies from Cisco Systems, Inc, VMware and EMC produced a highly efficient, robust and scalable Desktop Virtualization (DV) infrastructure for a hosted virtual desktop deployment. Key components of the solution included:
•
•
•
•
•
•
•
•
•
•
•
•
Architecture
Deployed Hardware
The architecture deployed is highly modular. While each customer's environment might vary in its exact configuration, once the reference architecture contained in this document is built, it can easily be scaled as requirements and demands change. This includes scaling both up (adding additional resources within a Cisco UCS Domain) and out (adding additional Cisco UCS Domains and VNX Storage arrays).
The 2000 User Horizon View 5.2 solution includes Cisco networking, Cisco Unified Computing System and EMC storage, all of which fits in two data center racks (one for the EMC VNX and one for the Cisco networking and Cisco UCS gear.) In fact, there is adequate rack space in the Cisco rack to add blades and chassis to support an additional 4000 users
This document details the deployment of VMware Horizon View 5.2 floating assignment linked clones on VMware ESXi 5.1. Cisco Nexus 1000V distributed switch manages the two VMware Clusters hosting the virtual desktops, insuring end-to-end Quality of Service and ease of management by the network team.
Figure 1 VMware Horizon View 5.2 2000 User Hardware Components
The reference configuration includes:
•
•
•
•
•
•
The EMC VNX5500 disk shelf, disk and Fast Cache configurations are detailed in section "Storage Architecture Design" later in this document.
Software Revisions
Table 1 Software Used in this Deployment
Configuration Guidelines
The 2000 User Horizon View 5.2 solution described in this document provides details for configuring a fully redundant, highly-available configuration. Configuration guidelines are provided that refer to which redundant component is being configured with each step, whether that be A or B. For example, SP A and SP B are used to identify the two EMC VNX storage controllers that are provisioned with this document while Nexus A and Nexus B identify the pair of Cisco Nexus switches that are configured. The Cisco UCS Fabric Interconnects are configured similarly.
This document is intended to allow the reader to configure the VMware Horizon View 5.2 customer environment as stand-alone solution.
VLANs
For the 2000 User Horizon View 5.2 solution, we utilized VLANs to isolate and apply access strategies to various types of network traffic. Table 2 details the VLANs used in this study.
Table 2 VLANs
VMware Clusters
Four VMware Clusters were used to support the solution and testing environment:
•
•
•
Infrastructure Components
This section describes all of the infrastructure components used in the solution outlined in this study.
Cisco Unified Computing System
Cisco Unified Computing System (UCS) is a set of pre-integrated data center components that comprises blade servers, adapters, fabric interconnects, and extenders that are integrated under a common embedded management system. This approach results in far fewer system components and much better manageability, operational efficiencies, and flexibility than comparable data center platforms.
Cisco Unified Computing System Components
Cisco UCS components are shown in Figure 2.
Figure 2 Cisco Unified Computing System Components
The Cisco Unified Computing System is designed from the ground up to be programmable and self-integrating. A server's entire hardware stack, ranging from server firmware and settings to network profiles, is configured through model-based management. With Cisco virtual interface cards, even the number and type of I/O interfaces is programmed dynamically, making every server ready to power any workload at any time.
With model-based management, administrators manipulate a model of a desired system configuration, associate a model's service profile with hardware resources and the system configures itself to match the model. This automation speeds provisioning and workload migration with accurate and rapid scalability. The result is increased IT staff productivity, improved compliance, and reduced risk of failures due to inconsistent configurations.
Cisco Fabric Extender technology reduces the number of system components to purchase, configure, manage, and maintain by condensing three network layers into one. It eliminates both blade server and hypervisor-based switches by connecting fabric interconnect ports directly to individual blade servers and virtual machines. Virtual networks are now managed exactly as physical networks are, but with massive scalability. This represents a radical simplification over traditional systems, reducing capital and operating costs while increasing business agility, simplifying and speeding deployment, and improving performance.
Fabric Interconnect
Cisco UCS Fabric Interconnects create a unified network fabric throughout the Cisco Unified Computing System. They provide uniform access to both networks and storage, eliminating the barriers to deploying a fully virtualized environment based on a flexible, programmable pool of resources.
Cisco Fabric Interconnects comprise a family of line-rate, low-latency, lossless 10-GE, Cisco Data Center Ethernet, and FCoE interconnect switches. Based on the same switching technology as the Cisco Nexus 5000 Series, Cisco UCS 6000 Series Fabric Interconnects provide the additional features and management capabilities that make them the central nervous system of Cisco Unified Computing System.
The Cisco UCS Manager software runs inside the Cisco UCS Fabric Interconnects. The Cisco UCS 6000 Series Fabric Interconnects expand the Cisco UCS networking portfolio and offer higher capacity, higher port density, and lower power consumption. These interconnects provide the management and communication backbone for the Cisco UCS B-Series Blades and Cisco UCS Blade Server Chassis.
All chassis and all blades that are attached to the Fabric Interconnects are part of a single, highly available management domain. By supporting unified fabric, the Cisco UCS 6200 Series provides the flexibility to support LAN and SAN connectivity for all blades within its domain right at configuration time. Typically deployed in redundant pairs, the Cisco UCS Fabric Interconnect provides uniform access to both networks and storage, facilitating a fully virtualized environment.
The Cisco UCS Fabric Interconnect family is currently comprised of the Cisco 6100 Series and Cisco 6200 Series of Fabric Interconnects.
Cisco UCS 6248UP 48-Port Fabric Interconnect
The Cisco UCS 6248UP 48-Port Fabric Interconnect is a 1 RU, 10-GE, Cisco Data Center Ethernet, FCoE interconnect providing more than 1Tbps throughput with low latency. It has 32 fixed ports of Fibre Channel, 10-GE, Cisco Data Center Ethernet, and FCoE SFP+ ports.
One expansion module slot can be up to sixteen additional ports of Fibre Channel, 10-GE, Cisco Data Center Ethernet, and FCoE SFP+.
Cisco UCS 6248UP 48-Port Fabric Interconnects were used in this study.
Cisco UCS 2200 Series IO Module
The Cisco UCS 2100/2200 Series FEX multiplexes and forwards all traffic from blade servers in a chassis to a parent Cisco UCS Fabric Interconnect over from 10-Gbps unified fabric links. All traffic, even traffic between blades on the same chassis, or VMs on the same blade, is forwarded to the parent interconnect, where network profiles are managed efficiently and effectively by the Fabric Interconnect. At the core of the Cisco UCS Fabric Extender are ASIC processors developed by Cisco that multiplex all traffic.
Note
Cisco UCS 2104 has eight 10GBASE-KR connections to the blade chassis mid-plane, with one connection per fabric extender for each of the chassis' eight half slots. This gives each half-slot blade server access to each of two 10-Gbps unified fabric-based networks via SFP+ sockets for both throughput and redundancy. It has 4 ports connecting up the fabric interconnect.
Cisco UCS 2208 has thirty-two 10GBASE-KR connections to the blade chassis midplane, with one connection per fabric extender for each of the chassis' eight half slots. This gives each half-slot blade server access to each of two 4x10-Gbps unified fabric-based networks via SFP+ sockets for both throughput and redundancy. It has 8 ports connecting up the fabric interconnect.
Note
Cisco UCS Chassis
The Cisco UCS 5108 Series Blade Server Chassis is a 6 RU blade chassis that will accept up to eight half-width Cisco UCS B-Series Blade Servers or up to four full-width Cisco UCS B-Series Blade Servers, or a combination of the two. The Cisco UCS 5108 Series Blade Server Chassis can accept four redundant power supplies with automatic load-sharing and failover and two Cisco UCS (either 2100 or 2200 series ) Fabric Extenders. The chassis is managed by Cisco UCS Chassis Management Controllers, which are mounted in the Cisco UCS Fabric Extenders and work in conjunction with the Cisco UCS Manager to control the chassis and its components.
A single Cisco UCS managed domain can theoretically scale to up to 40 individual chassis and 320 blade servers. At this time Cisco supports up to 20 individual chassis and 160 blade servers.
Basing the I/O infrastructure on a 10-Gbps unified network fabric allows the Cisco UCS to have a streamlined chassis with a simple yet comprehensive set of I/O options. The result is a chassis that has only five basic components:
•
•
•
•
•
Cisco UCS B200 M3 Blade Server
Cisco UCS B200 M3 is a third generation half-slot, two-socket Blade Server. The Cisco UCS B200 M3 harnesses the power of the latest Intel® Xeon® processor E5-2600 product family, with up to 384 GB of RAM (using 16-GB DIMMs), two optional SAS/SATA/SSD disk drives, and up to dual 4x 10 Gigabit Ethernet throughput, utilizing our VIC 1240 LAN on motherboard (LOM) design. The Cisco UCS B200 M3 further extends the capabilities of Cisco UCS by delivering new levels of manageability, performance, energy efficiency, reliability, security, and I/O bandwidth for enterprise-class virtualization and other mainstream data center workloads.
Cisco UCS VIC1240 Converged Network Adapter
A Cisco innovation, the Cisco UCS Virtual Interface Card (VIC) 1240 (Figure 1) is a 4-port 10 Gigabit Ethernet, Fibre Channel over Ethernet (FCoE)-capable modular LAN on motherboard (mLOM) designed exclusively for the M3 generation of Cisco UCS B-Series Blade Servers. When used in combination with an optional Port Expander, the Cisco UCS VIC 1240 capabilities can be expanded to eight ports of 10 Gigabit Ethernet.
The Cisco UCS VIC 1240 enables a policy-based, stateless, agile server infrastructure that can present up to 256 PCIe standards-compliant interfaces to the host that can be dynamically configured as either network interface cards (NICs) or host bus adapters (HBAs). In addition, the Cisco UCS VIC 1240 supports Cisco Data Center Virtual Machine Fabric Extender (VM-FEX) technology, which extends the Cisco UCS fabric interconnect ports to virtual machines, simplifying server virtualization deployment.
Figure 3 Cisco UCS VIC1240 Converged Network Adapter
Note
VMware Horizon View
VMware Horizon View (formerly known as VMware View) simplifies desktop and application management while increasing security and control. Horizon View delivers a personalized high fidelity experience for end-users across sessions and devices. It enables higher availability and agility of desktop services unmatched by traditional PCs while reducing the total cost of desktop ownership up to 50 percent. End-users can enjoy new levels of productivity and the freedom to access desktops from more devices and locations while giving IT greater policy control.
Horizon View 5.2 Features
Horizon View delivers rich, personalized virtual desktops as a managed service from a virtualization platform built to deliver the entire desktop, including the operating system, applications and data. With VMware Horizon View, desktop administrators virtualize the operating system, applications, and user data and deliver modern desktops to end-users. Get centralized automated management of these components for increased control and cost savings. Improve business agility while providing a flexible high performance desktop experience for end-users, across a variety of network conditions.
Automated Desktop Provisioning
Horizon View Manager provides a single management tool for greater IT efficiency to provision new desktops or groups of desktops, and an easy interface for setting desktop policies. Using a template, you can customize virtual pools of desktops and easily set policies, such as how many virtual machines can be in a pool, or logoff parameters.
Streamlined Application Management
VMware ThinApp application virtualization separates applications from underlying operating systems and reduces conflict between the OS and other applications for increased compatibility and streamlined management. Applications packaged with ThinApp can be run centrally from the datacenter, deployed locally to physical or virtual desktops or on USB drives for deployment flexibility.
Advanced Virtual Desktop Image Management
Horizon View Composer enables the rapid creation of desktop images from a golden image. Updates are instant and guaranteed across any number of virtual desktops. When combined with ThinApp, IT administrators can reduce the number of total images, storage requirements and operational costs.
Automate Desktop Operations Management
VMware vCenter Operations Manager for View allows administrators to gain insight into desktop and infrastructure performance, quickly pinpoint and troubleshoot issues. Administrators can optimize resource utilization, and proactively manage the desktop environment through the management dashboards. vCenter Operations Manager for View is an optional add-on for Horizon View customers. You can also leverage PCoIP Extension Services to collect Horizon View statistics into your existing WMI tool.
Efficient Resource Utilization
Horizon View Storage Accelerator optimizes storage load by caching common image blocks when reading virtual desktop images. Space Efficient Disks continuously reduce the storage needed per desktop. Both these technologies improve storage capacity and utilization, thereby reducing costs of additional hardware.
Built-in Security
Maintain control over data and intellectual property by keeping it secure in the datacenter. Encrypted protocol traffic provides secure end-users access virtual desktops inside or outside of the corporate network. Integration with vShield Endpoint enables offloaded and centralized anti-virus and anti-malware (AV) solutions. To eliminate agent sprawl and AV storm issues, risk of malware infection, and simplify AV administration. Horizon View also supports integration with Radius 2-factor authentication requirements.
Enhancements in Horizon View 5.2
VMware Horizon View 5.2 continues to build upon the advancements released in Horizon View 5.2. TCO was further reduced by optimizing storage reads, improved desktop migration and large scale management, and further enhanced the user-experience with lower bandwidth and client diversity.
Lower Total Cost of Ownership
Space-efficient disks, native in vSphere, reduce storage costs and administrative overhead by efficiently using and reclaiming storage space to minimize Horizon View Composer image size. This lowers storage capacity requirements for persistent desktops and decreases the need to continuously recompose and restore images.
Simplified Management
Improved large-scale management allows customers with large Horizon View deployments to efficiently and logically manage their virtual desktop infrastructure. Overall desktop architecture is simplified with a single VMware vCenter Server™ supporting up to 10,000 desktops in a pod. With support for 32 hosts per pool on VMFS along with NFS and pools spanning multiple VLANs, larger desktop pools can be created to decrease operational costs. Furthermore, View admin UI responsiveness increases and accelerates performance of operations such as provisioning, and rebalance improves the efficiency of the desktop administration team.
VMware vCenter Server virtual appliance support enables greater flexibility in Horizon View infrastructure deployment.
Seamless User Experience
Media services for rich 3D graphics add support for hardware accelerated 3D graphics for the most demanding 3D applications. By virtualizing the graphics processing unit (GPU), you can dedicate or share physical GPU resources across multiple users, providing a rich 3D experience from the data center. Using a combination of software and hardware-accelerated graphics, VMware Horizon View™ provides the greatest flexibility for delivering 3D graphics for virtual desktops and workstation use cases. 3D graphics acceleration is built upon the VMware vSphere® platform. Only Horizon View is designed to fully leverage vSphere, expanding the value of combined solutions.
Horizon View media services already support unified communications for Cisco Unified Communications.
Horizon View HTML access enables users to access desktops based on Horizon View from HTML5-capable browsers to securely access their data and applications. Without requiring the installation of any software or plug-ins, end users conveniently can access their desktops on any device. With Horizon Workspace™ integration, that same desktop convenience is expanded to provide end users access to their desktops, data and apps, all from a single location. Note: Horizon View HTML access is available in the Horizon View feature pack.
VMware Horizon View Clients for iOS and Android with Unity make it easier than ever to access Windows applications on your iPhone, iPad or Android device. Remove the frustration of working with Windows on mobile devices with a new mobile native user interface. With Unity users can easily browse, search, and open Windows applications and files, set applications and files as favorites, and easily switch between running applications.
Windows 8 support gives users ability to use the latest OS inside their virtual desktops. Horizon View Client has also been updated to run on the latest Windows 8 devices.
VMware Horizon View 5.2 Hosted Virtual Desktop Overview
Hosted Virtual Desktop (HVD) uses a hypervisor to host all the desktops in the data center.
Three types of HVD pools are available with Horizon View 5.2: Automated, Manual, and Terminal Services Pools. These pool types are discussed below.
•
•
•
•
For this study, we utilized Automated HVD pools with floating user assignments and Horizon View Composer linked clones over the PCoIP protocol.
Note
The following figure shows the logical architecture for a Horizon View 5.2 deployment, including the optional related product; Thin App. Thin App provides application streaming capability and is not included in this study.
Figure 4 VMware Horizon View 5.2 Architecture
EMC VNX Series
The VNX series delivers uncompromising scalability and flexibility for the mid-tier while providing market-leading simplicity and efficiency to minimize total cost of ownership. Customers can benefit from VNX features such as:
•
•
•
•
•
•
–
–
•
The VNX series includes five software suites and three software packs that make it easier and simpler to attain the maximum overall benefits.
•
–
–
–
–
–
•
–
–
–
EMC VNX5500 Used in Testing
EMC VNX 5500 is a unified storage platform for multi-protocol file, block and object storage. It is powered by Intel quad-core Xeon 5500 series processors and delivers five 9's availability. It is designed to deliver maximum performance and scalability for enterprise and mid-tier companies, enabling them to dramatically grow, share, and cost-effectively manage multi-protocol file and block systems. It supports up to 250 drives and three X-Blades (also known as Data Movers) for file protocol support.This solution was validated Fibre Channel for hypervisor SAN boot, data storage of virtual desktops, SQL database, and infrastructure virtual machines such Horizon View Connection Servers, Horizon View Composer Servers, VMware vCenter Servers, and other supporting services. An NFS or iSCSI variant could be deployed on the VNX5500 using NFS or iSCSI for data storage of virtual desktops.
VMware ESXi 5.1
VMware, Inc. provides virtualization software. VMware's enterprise software hypervisors for servers-VMware vSphere ESX, VMware vSphere ESXi, and VSphere-are bare-metal hypervisors that run directly on server hardware without requiring an additional underlying operating system.
VMware on ESXi 5.1 Hypervisor
ESXi 5.1 is a "bare-metal" hypervisor, so it installs directly on top of the physical server and partitions it into multiple virtual machines that can run simultaneously, sharing the physical resources of the underlying server. VMware introduced ESXi in 2007 to deliver industry-leading performance and scalability while setting a new bar for reliability, security and hypervisor management efficiency.
Due to its ultra-thin architecture with less than 100MB of code-base disk footprint, ESXi delivers industry-leading performance and scalability plus:
•
•
•
•
Modular Virtual Desktop Infrastructure Technical Overview
Modular Architecture
Today's IT departments are facing a rapidly-evolving workplace environment. The workforce is becoming increasingly diverse and geographically distributed and includes offshore contractors, distributed call center operations, knowledge and task workers, partners, consultants, and executives connecting from locations around the globe at all times.
An increasingly mobile workforce wants to use a growing array of client computing and mobile devices that they can choose based on personal preference. These trends are increasing pressure on IT to ensure protection of corporate data and to prevent data leakage or loss through any combination of user, endpoint device, and desktop access scenarios (Figure 5). These challenges are compounded by desktop refresh cycles to accommodate aging PCs and bounded local storage and migration to new operating systems, specifically Microsoft Windows 7.
Figure 5 The Evolving Workplace Landscape
Some of the key drivers for desktop virtualization are increased data security and reduced TCO through increased control and reduced management costs.
Cisco Data Center Infrastructure for Desktop Virtualization
Cisco focuses on three key elements to deliver the best desktop virtualization data center infrastructure: simplification, security, and scalability. The software combined with platform modularity provides a simplified, secure, and scalable desktop virtualization platform (Figure 6).
Figure 6 VMware Horizon View 5.2 on Cisco Unified Computing System
Simplified
Cisco Unified Computing System provides a radical new approach to industry standard computing and provides the heart of the data center infrastructure for desktop virtualization and the Cisco Virtualization Experience (VXI). Among the many features and benefits of Cisco Unified Computing System are the drastic reductions in the number of servers needed and number of cables per server and the ability to very quickly deploy or re-provision servers through Cisco UCS Service Profiles. With fewer servers and cables to manage and with streamlined server and virtual desktop provisioning, operations are significantly simplified. Thousands of desktops can be provisioned in minutes with Cisco Service Profiles and Cisco storage partners' storage-based cloning. This speeds time to productivity for end users, improves business agility, and allows IT resources to be allocated to other tasks.
IT tasks are further simplified through reduced management complexity, provided by the highly integrated Cisco UCS Manager, along with fewer servers, interfaces, and cables to manage and maintain. This is possible due to the industry-leading, highest virtual desktop density per blade of Cisco Unified Computing System along with the reduced cabling and port count due to the unified fabric and unified ports of Cisco Unified Computing System and desktop virtualization data center infrastructure.
Simplification also leads to improved and more rapid success of a desktop virtualization implementation. Cisco and its partners -VMware and EMC - have developed integrated, validated architectures, including available pre-defined, validated infrastructure packages, known as Cisco Solutions for EMC VSPEX End User Computing.
Secure
While virtual desktops are inherently more secure than their physical world predecessors, they introduce new security considerations. Desktop virtualization significantly increases the need for virtual machine-level awareness of policy and security, especially given the dynamic and fluid nature of virtual machine mobility across an extended computing infrastructure. The ease with which new virtual desktops can proliferate magnifies the importance of a virtualization-aware network and security infrastructure. Cisco Unified Computing System and Nexus data center infrastructure for desktop virtualization provides stronger data center, network, and desktop security with comprehensive security from the desktop to the hypervisor. Security is enhanced with segmentation of virtual desktops, virtual machine-aware policies and administration, and network security across the LAN and WAN infrastructure.
Scalable
Growth of a desktop virtualization solution is all but inevitable and it is critical to have a solution that can scale predictably with that growth. The Cisco solution supports more virtual desktops per server and additional servers scale with near linear performance. Cisco data center infrastructure provides a flexible platform for growth and improves business agility. Cisco UCS Service Profiles allow for on-demand desktop provisioning, making it easy to deploy dozens or thousands of additional desktops.
Each additional Cisco UCS server provides near linear performance and utilizes Cisco's dense memory servers and unified fabric to avoid desktop virtualization bottlenecks. The high performance, low latency network supports high volumes of virtual desktop traffic, including high resolution video and communications.
Cisco Unified Computing System and Nexus data center infrastructure is an ideal platform for growth, with transparent scaling of server, network, and storage resources to support desktop virtualization.
Savings and Success
As demonstrated above, the simplified, secure, scalable Cisco data center infrastructure solution for desktop virtualization will save time and cost. There will be faster payback, better ROI, and lower TCO with the industry's highest virtual desktop density per server resulting in fewer servers required, reducing both capital expenditures (CapEx) and operating expenditures (OpEx). There will also be much lower network infrastructure costs, with fewer cables per server and fewer ports required, via the Cisco UCS architecture and unified fabric.
The simplified deployment of Cisco Unified Computing System for desktop virtualization speeds up time to productivity and enhances business agility. IT staff and end users are more productive more quickly and the business can react to new opportunities by simply deploying virtual desktops whenever and wherever they are needed. The high performance Cisco systems and network deliver a near-native end-user experience, allowing users to be productive anytime, anywhere.
Understanding Desktop User Groups
There must be a considerable effort within the enterprise to identify desktop user groups and their memberships. The most broadly recognized, high-level user groups are as follows:
•
•
•
•
•
•
There is good reason to search for and identify multiple sub-groups of the major groups listed above in the enterprise. Typically, each sub-group has different application and data requirements.
Understanding Applications and Data
When the desktop user groups and sub-groups have been identified, the next task is to catalog group application and data requirements. This can be one of the most time-consuming processes in the VDI planning exercise, but is essential for the VDI project's success. If the applications and data are not identified and co-located, performance will be negatively affected.
The process of analyzing the variety of application and data pairs for an organization will likely be complicated by the inclusion cloud applications, like SalesForce.com. This application and data analysis is beyond the scope of this Cisco Validated Design, but should not be omitted from the planning process. There are a variety of third party tools available to assist organizations with this crucial exercise.
Project Planning and Solution Sizing Sample Questions
Now that user groups, their applications and their data requirements are understood, some key project and solution sizing questions may be considered.
General project questions should be addressed at the outset, including:
•
•
•
•
•
•
Provided below is a short, non-exhaustive list of sizing questions that should be addressed for each user sub-group:
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Proof of Concept Pilot Program
To validate what you have learned during your analysis, create an isolated Proof of Concept environment to test the various workloads and validate your sizing calculations.
Then create a Pilot environment and get users from each user group who can exercise all of the organizations key applications. Use the pilot user feedback to further refine you design in preparation for production roll out.
Failure to follow these key steps will make a successful virtual desktop deployment project nearly impossible.
Cisco Services
Cisco offers assistance for customers in the analysis, planning, implementation, and support phases of the VDI lifecycle. These services are provided by the Cisco Advanced Services group. Some examples of Cisco services include:
•
•
•
The Solution: A Unified, Pre-Tested and Validated Infrastructure
To meet the challenges of designing and implementing a modular desktop infrastructure, Cisco, EMC and VMware have collaborated to create the data center solution for virtual desktops outlined in this document.
Key elements of the solution include:
•
•
Cisco Networking Infrastructure
This section describes the Cisco networking infrastructure components used in the configuration.
Cisco Nexus 5548 Switch
The Cisco Nexus 5548UP is a 1-RU 10 Gigabit Ethernet, Fibre Channel, and FCoE switch offering up to 960 Gbps of throughput and up to 48 ports. The switch has 32 unified ports that accept modules and cables meeting the Small Form-Factor Pluggable Plus (SFP+) standard and one expansion slot. Expansion slot options include:
•
•
•
•
The switch has a single serial console port and a single out-of-band 10/100/1000-Mbps Ethernet management port. Two N+1 redundant, hot-pluggable power supplies and five N+1 redundant, hot-pluggable fan modules provide highly reliable front-to-back cooling.
Cisco Nexus 5500 Series Feature Highlights
The switch family's rich feature set makes the series ideal for rack-level, access-layer applications. It protects investments in data center racks with standards-based Ethernet and FCoE features that allow IT departments to consolidate networks based on their own requirements and timing.
•
•
•
Features and Benefits
This sections details the specific features and benefits provided by the Cisco Nexus 5500 Series.
10GB Ethernet, FCoE, and Unified Fabric Features
The switch series, using cut-through architecture, supports line-rate 10 Gigabit Ethernet on all ports while maintaining consistently low latency independent of packet size and services enabled. It supports a set of network technologies known collectively as Data Center Bridging (DCB) that increases the reliability, efficiency, and scalability of Ethernet networks. These features allow the switches to support multiple traffic classes over a lossless Ethernet fabric, thus enabling consolidation of LAN, SAN, and cluster environments. Its ability to connect Fibre Channel over Ethernet (FCoE) to native Fibre Channel protects existing storage system investments while dramatically simplifying in-rack cabling.
Low Latency
The cut-through switching technology used in the Cisco Nexus 5500 Series ASICs enables the product to offer a low latency of 3.2 microseconds, which remains constant regardless of the size of the packet being switched. This latency was measured on fully configured interfaces, with access control lists (ACLs), QoS, and all other data path features turned on. The low latency on the Cisco Nexus 5500 Series enables application-to-application latency on the order of 10 microseconds (depending on the NIC). These numbers, together with the congestion management features described in the next section, make the Cisco Nexus 5500 Series a great choice for latency-sensitive environments.
Other Features
Other Nexus 5548UP features include: Nonblocking Line-Rate Performance, Single-Stage Fabric, Congestion Management, Virtual Output Queues, Lossless Ethernet (Priority Flow Control), Delayed Drop FC over Ethernet, Hardware-Level I/O Consolidation, and End-Port Virtualization.
Cisco Nexus 1000V Feature Highlight
Cisco Nexus 1000V Series Switches are virtual machine access switches that are an intelligent software switch implementation based on IEEE 802.1Q standard for VMware vSphere environments running the Cisco® NX-OS Software operating system. Operating inside the VMware ESX hypervisor, the Cisco Nexus 1000V Series supports Cisco VN-Link server virtualization technology to provide:
•
•
•
With the Cisco Nexus 1000V Series, you can have a consistent networking feature set and provisioning process all the way from the virtual machine access layer to the core of the data center network infrastructure. Virtual servers can now use the same network configuration, security policy, diagnostic tools, and operational models as their physical server counterparts attached to dedicated physical network ports. Virtualization administrators can access pre-defined network policy that follows mobile virtual machines to help ensure proper connectivity, saving valuable time for virtual machine administration.
Developed in close collaboration with VMware, the Cisco Nexus 1000V Series is certified by VMware to be compatible with VMware vSphere, vCenter, ESX, and ESXi, and with many other vSphere features. You can use the Cisco Nexus 1000V Series to manage your virtual machine connectivity with confidence in the integrity of the server virtualization infrastructure.
The Cisco Nexus 1000V Release 2.1 software is being offered in two editions:
•
•
Cisco Nexus 1000V Product Architecture
Cisco Nexus 1000V Series Switches have two major components: the Virtual Ethernet Module (VEM), which runs inside the hypervisor, and the external Virtual Supervisor Module (VSM), which manages the VEMs.
Virtual Ethernet Module (VEM)
The Cisco Nexus 1000V Series VEM runs as part of the VMware ESX or ESXi kernel and replaces the VMware virtual switch (vSwitch). This level of integration helps ensure that the Cisco Nexus 1000V Series is fully aware of all server virtualization events, such as VMware vMotion and Distributed Resource Scheduler (DRS). The VEM takes configuration information from the VSM and provides advanced networking functions: quality of service (QoS), security features, and monitoring features.
Virtual Supervisor Module (VSM)
The Cisco Nexus 1000V Series VSM controls multiple VEMs as one logical modular switch. Configuration is performed through the VSM and is automatically propagated to the VEMs. Instead of configuring soft switches inside the hypervisor on a host-by-host basis administrators can define configurations for immediate use on all VEMs being managed by the VSM from a single interface.
Cisco Nexus 1000V Features and Benefits
The Cisco Nexus 1000V Series provides a common management model for both physical and virtual network infrastructures through Cisco VN-Link technology, which includes policy-based virtual machine connectivity, mobility of virtual machine security and network properties, and a non-disruptive operational model.
Policy-Based Virtual Machine Connectivity
To facilitate easy creation and provisioning of virtual machines, the Cisco Nexus 1000V Series includes port profiles. Port profiles enable you to define virtual machine network policies for different types or classes of virtual machines and then apply the profiles through the VMware vCenter. Port profiles are a scalable mechanism for configuring networks with large numbers of virtual machines. When the Port Profiles include QoS and security policies, they formulate a complete service-level agreement (SLA) for the virtual machine's traffic.
Mobility of Virtual Machine Security and Network Properties
Network and security policies defined in the port profile follow the virtual machine throughout its lifecycle, whether it is being migrated from one server to another, suspended, hibernated, or restarted. In addition to migrating the policy, the Cisco Nexus 1000V Series VSM moves the virtual machine's network state. Virtual machines participating in traffic-monitoring activities can continue these activities uninterrupted by VMware vMotion operations. When a specific port profile is updated, the Cisco Nexus 1000V Series automatically provides live updates to all the virtual ports using that same port profile. The capability to migrate network and security policies through VMware vMotion makes regulatory compliance much easier to enforce with the Cisco Nexus 1000V Series because the security policy is defined in the same way as for physical servers and is constantly enforced by the switch.
Besides traditional switching capability, the Cisco Nexus 1000V Series offers the Cisco vPath architecture to support virtualized network services with:
•
•
Cisco Virtual Service Gateway (VSG) is the first VSN to leverage the Cisco vPath architecture and provides multi-tenant, scalable, security services for virtual machines on the Cisco Nexus 1000V Series Switches.
Non-disruptive Operational Model
Because of its close integration with VMware vCenter, the Cisco Nexus 1000V Series allows virtualization administrators to continue using VMware tools to provision virtual machines. At the same time, network administrators can provision and operate the virtual machine network the same way they do the physical network. While both teams work independently, the Cisco Nexus 1000V Series enforces consistent configuration and policy throughout the server virtualization environment. This level of integration lowers the cost of ownership while supporting organizational boundaries among server, network, security, and storage teams.
Inside VMware vCenter, virtual machines are configured as before. For network configuration, port profiles defined on the Cisco Nexus 1000V Series VSM are displayed by VMware vCenter as port groups. Virtualization administrators can take advantage of preconfigured port groups and focus on virtual machine management, and network administrators can use port profiles to apply policy for a large number of ports at the same time. Together, both teams can deploy server virtualization more efficiently and with lower operating costs.
Enhanced Deployment Scenarios
•
•
•
VMware Product Compatibility
The Cisco Nexus 1000V Series is compatible with VMware vSphere as a VMware vNetwork Distributed Switch (vDS) with support for VMware ESX and ESXi hypervisors and integration with VMware vCenter Server. Cisco Nexus 1000V Series Switches are compatible with the various VMware vSphere features.
Architecture and Design of Horizon View 5.2 on Cisco Unified Computing System and EMC VNX Storage
Design Fundamentals
There are many reasons to consider a virtual desktop solution such as an ever growing and diverse base of user devices, complexity in management of traditional desktops, security, and even Bring Your Own Computer (BYOC) to work programs. The first step in designing a virtual desktop solution is to understand the user community and the type of tasks that are required to successfully execute their role. The following user classifications are provided:
•
•
•
•
•
After the user classifications have been identified and the business requirements for each user classification have been defined, it becomes essential to evaluate the types of virtual desktops that are needed based on user requirements. There are essentially five potential desktops environments for each user:
•
•
•
•
•
For the purposes of the validation represented in this document only hosted virtual desktops were validated. Each of the sections provides some fundamental design decisions for this environment.
Hosted Virtual Desktop Design Fundamentals
This section details how the VMware Horizon View 5.2 can be used to deliver a variety of virtual desktop configurations.
Choosing a Display Protocol
A display protocol provides end users with a graphical interface to a View desktop that resides in the datacenter. You can use PCoIP (PC-over-IP), which VMware provides, or Microsoft RDP (Remote Desktop Protocol.)
You can set policies to control which protocol is used or to allow end users to choose the protocol when they login to a desktop.
Note
VMware View with PCoIP
PCoIP provides an optimized desktop experience for the delivery of the entire desktop environment, including applications, images, audio, and video content for a wide range of users on the LAN or across the WAN. PCoIP can compensate for an increase in latency or a reduction in bandwidth, to make sure that end users can remain productive regardless of network conditions.
PCoIP is supported as the display protocol for View desktops with virtual machines and with physical machines that contain Teradici host cards.
PCoIP Features
Key features of PCoIP include the following:
•
•
•
•
•
•
•
•
•
•
Video Quality
•
•
•
•
Recommended Guest Settings
Recommended guest operating system settings include the following settings:
•
•
Microsoft RDP
Remote Desktop Protocol is the same multichannel protocol many people already use to access their work computer from their home computer. Microsoft Remote Desktop Connection (RDC) uses RDP to transmit data.
Microsoft RDP provides the following features:
•
•
•
•
•
•
•
•
•
•
•
Note
Recommended Guest Settings
Client hardware requirements include the following:
•
•
•
Choose a User Profile Management System
There are a number of options for managing user profiles for HVDs. The two methods we considered for this study were Microsoft Roaming User Profiles and View Persona Manager. It is important to select and deploy a method so that user settings for software applications and user preferences are maintained, particularly for floating desktops. Both methods are discussed briefly below. (We used Microsoft Roaming User Profiles in the study.)
Microsoft Roaming User Profiles and Folder Redirection
This technology has been around for more than a dozen years. It was significantly enhanced with the introduction of Windows Vista and updated again with Windows 7. Version two (v2) roaming profiles were introduced, adding 8 additional folders that can be redirected. This greatly reduces the time it takes to load the user's profile during logon.Using Roaming User Profiles and Folder redirection require a network shares that all users have access to during the virtual desktop session. The user must have read and write access to their profile folder and folder redirection folder, which get created on first login after Roaming User Profiles is configured.
Utilizing Microsoft Active Directory Group Policy is the recommended method for providing Roaming User Profiles and Folder Redirection to your users. See the article titled Managing Roaming User Data Deployment Guide at the following url for details on how to configure both Roaming User Profiles and Folder Redirection:
http://technet.microsoft.com/en-us/library/cc766489%28WS.10%29.aspx
Note
VMware Persona Management
You can use View Persona Management with View desktops on physical computers and virtual machines that are not managed by View. View Persona Management retains changes that users make to their profiles. User profiles comprise a variety of user-generated information.
•
•
•
To facilitate these abilities, View Persona Management requires storage on a CIFS share equal or greater than the size of the user's local profile.
Minimizing Logon and Logoff Times
View Persona Management minimizes the time it takes to log on to and off of desktops.
•
•
•
With View Persona Management, you can avoid making any changes to Active Directory in order to have a managed profile. To configure Persona Management, you specify a central repository, without changing the user's properties in Active Directory. With this central repository, you can manage a user's profile in one environment without affecting the physical machines that users might also log on to.
With View Persona Management, if you provision desktops with VMware ThinApp applications, the ThinApp sandbox data can also be stored in the user profile. This data can roam with the user but does not significantly affect logon times. This strategy provides better protection against data loss or corruption.
Configuration Options
You can configure View personas at several levels: a single View desktop, a desktop pool, an OU, or all View desktops in your deployment. You can also use a standalone version of View Persona Management on physical computers and virtual machines that are not managed by View.
By setting group policies (GPOs), you have granular control of the files and folders to include in a persona:
•
•
•
•
As with Windows roaming profiles, you can configure folder redirection. You can redirect the same folders that support redirection with Windows Roaming User Profiles.
Accessing USB Devices Connected to the End Point
Administrators can configure the ability to use USB devices, such as thumb flash drives, VoIP (voice-over-IP) devices, and printers, from a View desktop. This feature is called USB redirection. (It was not used in this study.)
When you use this feature, most USB devices that are attached to the local client system become available from a menu in View Client. You use the menu to connect and disconnect the devices.
You can specify which types of USB devices end users are allowed to connect to. For composite devices that contain multiple types of devices, such as a video input device and a storage device, you can split the device so that one device (for example, the video input device) is allowed but the other device (for example, the storage device) is not.
USB devices that do not appear in the menu, but are available in a View desktop, include smart card readers and human interface devices such as keyboards and pointing devices. The View desktop and the local computer use these devices at the same time.
This feature has the following limitations:
•
•
Printing from a View Desktop
The virtual printing feature allows end users with View Client on Windows systems to use local or network printers from a View desktop without requiring that additional print drivers be installed in the View desktop.
The location-based printing feature allows you to map View desktops to the printer that is closest to the endpoint client device.
With virtual printing, after a printer is added on a local Windows computer, View adds that printer to the list of available printers on the View desktop. No further configuration is required. For each printer available through this feature, you can set preferences for data compression, print quality, double-sided printing, color, and so on. Users who have administrator privileges can still install printer drivers on the View desktop without creating a conflict with the virtual printing component. To send print jobs to a USB printer, you can either use the USB redirection feature or use the virtual printing feature.
The location-based printing feature is available for both Windows and non-Windows client systems. Location based printing allows IT organizations to map View desktops to the printer that is closest to the endpoint client device. Using this feature does require that the correct printer drivers be installed in the View desktop.
We did not use virtual printing in this study. Our workload generator, Login VSI, installs a pdf printer into the master image which is utilized for printing during the test.
Other Features to Consider
Horizon View 5.2 supports these additional features that were not deployed in this study:
•
•
•
Designing a VMware Horizon View 5.2 Deployment
There are several elements that go into the design of a successful Horizon View 5.2 environment. This section covers those topics at a high level. Readers should consult the VMware View Architecture Planning guide for Horizon View 5.2 at the following URL for more details: https://pubs.vmware.com/view-52/topic/com.vmware.ICbase/PDF/horizon-view-52-architecture-planning.pdf
Determine Desktop Pools Required
Based on the analysis performed on user groups and the applications identified that will be supported by the Hosted Virtual Desktop (HVD) environment, a strategy for laying out your desktop pool structure should be create.
For this study, a single user group (knowledge workers) will be tested and the application workload that this group will run will be identified, which is based on the Login VSI 3.6 medium workload (with flash.) Virtual machines will be used as the desktop sourc.
If you use a vSphere virtual machine as a desktop source, you can automate the process of making as many identical virtual desktops as you need. You can set a minimum and maximum number of virtual desktops to be generated for the pool. Setting these parameters ensures that you always have enough View desktops available for immediate use but not so many that you overuse available resources.
Using pools to manage desktops allows you to apply settings or deploy applications to all virtual desktops in a pool. The following examples show some of the settings available:
•
•
•
•
•
In addition, using desktop pools provides many conveniences.
•
•
Using floating-assignment pools also allows you to create a pool of desktops that can be used by shifts of users. For example, a pool of 100 desktops could be used by 300 users if they worked in shifts of 100 users at a time.
Note
Managing Storage Requirements
VMware vSphere lets you virtualize disk volumes and file systems so that you can manage and configure storage without having to consider where the data is physically stored.
Fibre Channel SAN arrays, iSCSI SAN arrays, and NAS arrays are widely used storage technologies supported by VMware vSphere to meet different datacenter storage needs. The storage arrays are connected to and shared between groups of servers through storage area networks. This arrangement allows aggregation of the storage resources and provides more flexibility in provisioning them to virtual machines.
With View 4.5 and later and vSphere 4.1 and later, you can now also use the following features:
•
•
•
With Horizon View 5.2 and later and vSphere 5.0 and later, you can now also use the following features:
•
Using this content-based read cache (CBRC) can reduce IOPS and improve performance during boot storms, when many desktops start up and run anti-virus scans at the same time. Instead of reading the entire OS from the storage system over and over, a host can read common data blocks from cache.
•
Although replica disks must be stored on NFS datastores, OS disks and persistent disks can be stored on NFS or VMFS datastores.
View Composer
Because View Composer creates desktop images that share virtual disks with a base image, you can reduce the required storage capacity by 50 to 90 percent.
View Composer uses a base image, or parent virtual machine, and creates a pool of up to 1,000 linked-clone virtual machines. Each linked clone acts like an independent desktop, with a unique host name and IP address, yet the linked clone requires significantly less storage.
When creating a linked-clone desktop pool, a full clone is first made from the parent virtual machine. The full clone, or replica, and the clones linked to it can be placed in a variety of locations. The options are:
•
•
As an example, you could place the replicas on low capacity read optimized drives with IOPS and place the linked clones on traditional spinning media
•
Guest OS page files and temp files are placed here. When the HVD is powered off, this disk is deleted
•
End user's application data and profiles are stored here. The data survives refresh, recompose and rebalance operations.
•
Host local drives store linked clone files, presenting some advantages and several disadvantages. Use this option with care after considering your requirements.
Note
Hosted Virtual Desktop Infrastructure Design
To implement the automated pool floating desktop delivery model for this study, the VMware View Reference Architecture for virtual desktop delivery was followed.
Figure 7 View Desktop Infrastructure
Learn more about VMware Horizon View planning and design at the following location:
Solution Validation
This section details the configuration and tuning that was performed on the individual components to produce a complete, validated solution.
Configuration Topology for Scalable VMware View 5.2 Virtual Desktop Infrastructure on Cisco Unified Computing System and EMC Storage
Figure 8 illustrates the Cisco Unified Computing System VDI configuration.
Figure 8 Architecture Block
Figure 8 illustrates the architectural topology for the purpose of this study. The architecture is divided into four distinct layers:
•
•
•
•
Figure 9 details the physical configuration of the 2000 seat View 5.2 environment.
Figure 9 Detailed Architecture Configuration
Cisco Unified Computing System Configuration
This section talks about the Cisco UCS configuration that was done as part of the infrastructure build out. The racking, power and installation of the chassis are described in the install guide (see http://www.cisco.com/en/US/docs/unified_computing/ucs/hw/chassis/install/ucs5108_install.html) and it is beyond the scope of this document. More details on each step can be found in the following documents:
•
•
Base Cisco Unified Computing System Configuration
To configure the Cisco Unified Computing System, perform the following steps:
1.
2.
3.
If the firmware is not current, follow the installation and upgrade guide to upgrade the Cisco UCS Manager firmware. We will use Cisco UCS Policy in Service Profiles later in this document to update all Cisco UCS components in the solution.
Note
4.
5.
6.
Use the Configure Unified Ports, Configure Expansion Module Ports to configure FC uplinks.
Note
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
Note
21.
22.
23.
24.
The Advanced Tab Settings:
The remaining Advanced tab settings are at the platform default or not configured. Similarly, the Boot Options and Server Management tabs' settings are set at the defaults.
Note
25.
Note
Note
26.
27.
Note
Follow through each section, utilizing the policies and objects you created earlier, then click Finish.
Note
Note
28.
29.
Note
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
Note
44.
45.
46.
Note
47.
The image below is a Cisco UCS B200 M3 Sample:
The Cisco UCS Blade Servers are ready for the hypervisor installation.
QoS and CoS in Cisco Unified Computing System
Cisco Unified Computing System provides different system class of service to implement quality of service including:
•
•
•
Applications like the Cisco Unified Computing System and other time sensitive applications have to adhere to a strict QOS for optimal performance.
System Class Configuration
Systems Class is the global operation where entire system interfaces are with defined QoS rules.
•
Best effort is equivalent in MQC terminology as "match any"
–
•
–
–
–
–
•
•
•
•
(Weight of the given priority * 100)% b/w shared of given Class = ________________________________Sum of weights of all priority
Cisco UCS Class Configuration
•
•
•
•
Table 3 Name Table Map Between Cisco UCS and the NXOS
Table 4 Class to CoS Map by default in Cisco Unified Computing System
Table 5 Default Weight in Cisco Unified Computing System
Steps to Enable QOS on Cisco Unified Computing System
For this study, we utilized four Cisco UCS QoS System Classes to priorities four types of traffic in the infrastructure.
Table 6 QoS Priority to vNIC and VLAN Mapping
Configure Platinum, Gold, Silver and Bronze policies by checking the enabled box. The Platinum Policy, used for NFS storage, was configured for Jumbo Frames in the MTU column. Notice the option to set no packet drop policy during this configuration.
Figure 10 Cisco UCS QoS System Class Configuration
Next, from the LAN tab under Policies, Root, QoS Polices, verify QoS Policies Platinum, Gold, Silver and Bronze exist, with each QoS policy mapped to its corresponding Priority.
Figure 11 Cisco UCS QoS Policy Configuration
Finally, include the corresponding QoS Policy into each vNIC template using the QoS policy drop down, using the QoS Priority to vNIC and VLAN Mapping table above.
Figure 12 Utilize QoS Policy in vNIC Template
This is a unique value proposition for Cisco Unified Computing System with respect to end-to-end QOS. For example, there is a VLAN for the EMC storage; configure the Platinum policy with Jumbo frames and receive an end-to-end QOS and performance guarantees from the Blade Servers to the Nexus 1000V virtual distributed switches running in vCenter through the Nexus 5548UP access layer switches.
LAN Configuration
The access layer LAN configuration consists of a pair of Cisco Nexus 5548s (N5Ks,) a family member of our low-latency, line-rate, 10 Gigabit Ethernet and FCoE switches for our VDI deployment.
Cisco UCS Connectivity
Four 10 Gigabit Ethernet uplink ports are configured on each of the Cisco UCS 6248 fabric interconnects, and they are connected to the Cisco Nexus 5548 pair in a bow tie manner as shown below in a port channel.
The 6248 Fabric Interconnect is in End host mode, as we are doing both Fiber Channel as well as Ethernet (NAS) data access as per the recommended best practice of the Cisco Unified Computing System. We built this out for scale and have provisioned more than 40 G per Fabric interconnect (Figure 13).
Note
Figure 13 Ethernet Network Configuration with Upstream Cisco Nexus 5500 Series from the Cisco Unified Computing System 6200 Series Fabric Interconnects
EMC VNX5500 LAN Connectivity
The Cisco Nexus 5548UP is used to connect to the EMC VNX5500 storage system for Fiber Channel and file-based access.
The VNX5500 is equipped with dual-port 8GB FC modules on each controller. These are connected to the pair of Nexus 5548 unified ports to provide block storage access to the environment. (See section SAN Configuration)
The VNX5500 supports two dual-port 10G Data Movers which are connected to the pair of N5Ks downstream. One of the Data Movers is set to Active, with the second providing failover capability. This allows end-to-end 10G access for file-based storage traffic. We have implemented jumbo frames on the ports and have priority flow control on, with Platinum CoS and QoS assigned to the vNICs carrying the storage data access on the Fabric Interconnects. (Note: This configuration was not used in this study, but is shown as a supported option.)
The EMC ethernet connectivity diagram is shown below. There is a total of 40 Gbps bandwidth available for the servers.
Figure 14 EMC VNX Ethernet Connectivity
Figure 15 EMC VNX Fibre Channel Connectivity
For information on configuring ethernet connectivity on a EMC VNX5500 Storage System, refer to the EMC website: www.emc.com.
Nexus 1000V Configuration in L3 Mode
1.
2.
3.
4.
5.
6.
7.
8.
9.
a.
b.
c.
10.
11.
12.
13.
14.
VDI-N1KV# conf t15.
VDI-N1KV(config)# vlan 122VDI-N1KV(config-vlan)# name VDAVDI-N1KV(config-vlan)# no shVDI-N1KV(config)# vlan 164VDI-N1KV(config-vlan)# name ESXi-MgmtVDI-N1KV(config-vlan)# no shVDI-N1KV(config)# vlan 165VDI-N1KV(config-vlan)# name Infra-MgmtVDI-N1KV(config-vlan)# no shVDI-N1KV(config)# vlan 166VDI-N1KV(config-vlan)# name StorageVDI-N1KV(config-vlan)# no shVDI-N1KV(config)# vlan 167VDI-N1KV(config-vlan)# name N1K-ControlVDI-N1KV(config-vlan)# no shVDI-N1KV(config)# vlan 169VDI-N1KV(config-vlan)# name vMotionVDI-N1KV(config-vlan)# no shVDI-N1KV(config)#<CNTRL/Z>
16.
VDI-N1KV# conf tVDI-N1KV(config)# policy-map type qos jumbo-mtuVDI-N1KV(config-pmap-qos)# policy-map type qos platinum_Cos_5VDI-N1KV(config-pmap-qos)# class class-defaultVDI-N1KV(config-pmap-c-qos)# set cos 5VDI-N1KV(config-pmap-c-qos)# endVDI-N1KV# copy running-config startup-config
17.
Note
18.
VDI-N1KV(config)# port-profile type ethernet Mgmt-uplinkVDI-N1KV(config)# vmware port-groupVDI-N1KV(config-port-prof)# switchport mode trunkVDI-N1KV(config-port-prof)# switchport trunk allowed vlan 164-165,167VDI-N1KV(config-port-prof)# channel-group auto mode on mac-pinningVDI-N1KV(config-port-prof)# no shutdownVDI-N1KV(config-port-prof)# system vlan 164,167VDI-N1KV(config-port-prof)#state enabled19.
VDI-N1KV(config)# port-profile type ethernet storage-uplinkVDI-N1KV(config)# vmware port-groupVDI-N1KV(config-port-prof)# switchport mode accessVDI-N1KV(config-port-prof)# switchport access vlan 166VDI-N1KV(config-port-prof)# mtu 9000VDI-N1KV(config-port-prof)# channel-group auto mode on mac-pinningVDI-N1KV(config-port-prof)# no shutdownVDI-N1KV(config-port-prof)# system vlan 166VDI-N1KV(config-port-prof)#state enabled
20.
VDI-N1KV(config)# port-profile type vethernet storageVDI-N1KV(config-port-prof)# vmware port-groupVDI-N1KV(config-port-prof)# switchport mode accessVDI-N1KV(config-port-prof)# switchport access vlan 166VDI-N1KV(config-port-prof)# service -policy type qos input platinum_Cos_5VDI-N1KV(config-port-prof)# no shVDI-N1KV(config-port-prof)# system vlan 166VDI-N1KV(config-port-prof)#state enabled
21.
VDI-N1KV(config)# port-profile type ethernet vmotion-uplinkVDI-N1KV(config-port-prof)# vmware port-groupVDI-N1KV(config-port-prof)# switchport mode accessVDI-N1KV(config-port-prof)# switchport access vlan 169VDI-N1KV(config-port-prof)# channel-group auto mode on mac-pinningVDI-N1KV(config-port-prof)# no shVDI-N1KV(config-port-prof)# system vlan 169VDI-N1KV(config-port-prof)#state enabled
22.
VDI-N1KV(config)# port-profile type vethernet vmotionVDI-N1KV(config-port-prof)# vmware port-groupVDI-N1KV(config-port-prof)# switchport mode accessVDI-N1KV(config-port-prof)# switchport access vlan 169VDI-N1KV(config-port-prof)# no shVDI-N1KV(config-port-prof)# system vlan 169VDI-N1KV(config-port-prof)#state enabled
23.
VDI-N1KV(config)# port-profile type ethernet vdi-uplinkVDI-N1KV(config-port-prof)# vmware port-groupVDI-N1KV(config-port-prof)#switchport mode accessVDI-N1KV(config-port-prof)#switchport access vlan 122,164-165VDI-N1KV(config-port-prof)#channel-group auto mode on mac-pinningVDI-N1KV(config-port-prof)#no shutdownVDI-N1KV(config-port-prof)# system vlan 122,165VDI-N1KV(config-port-prof)#state enabled
24.
VDI-N1KV(config)# port-profile type vethernet vdi-pool1VDI-N1KV(config)# vmware port-groupVDI-N1KV(config-port-prof)# port-binding static auto expandVDI-N1KV(config-port-prof)# max-ports 1024VDI-N1KV(config-port-prof)# switchport mode accessVDI-N1KV(config-port-prof)# switchport access vlan 122VDI-N1KV(config-port-prof)# no shVDI-N1KV(config-port-prof)# system vlan 122VDI-N1KV(config-port-prof)#state enabled
25.
VDI-N1KV(config)# port-profile type vethernet vdi-pool2VDI-N1KV (config-port-prof)# vmware port-groupVDI-N1KV(config-port-prof)# port-binding static auto expandVDI-N1KV (config-port-prof)# max-ports 1024VDI-N1KV(config-port-prof)# switchport mode accessVDI-N1KV(config-port-prof)# switchport access vlan 122VDI-N1KV(config-port-prof)# no shVDI-N1KV (config-port-prof)# system vlan 122VDI-N1KV(config-port-prof)#state enable
26.
27.
28.
Note
29.
30.
For example, consider vmnic0 and vmnic1 for use as the System-Uplink. As per the best practices here we have 8 vmnics (4 pairs) and each pair of vmnics will be associated with one uplink; system/mgmt uplink, storage uplink, VM/VDI traffic uplink, vMotion Uplink.
31.
32.
33.
34.
35.
Note
In the absence of Update manager:
1.
2.
3.
esxcli software vib install -v /vmfs/volumes/datastore/ cross_cisco-vem-v144-4.2.1.1.5.2.0-3.0.1.vib4.
5.
VDI-N1KV(config)# sh module
6.
SAN Configuration
The same pair of Nexus 5548UP switches were used in the configuration to connect between the FC ports on the EMC VNX5500 and the FC ports of the UCS 6248 Fabric Interconnects.
Boot from SAN Benefits
Booting from SAN is another key feature which helps in moving towards stateless computing in which there is no static binding between a physical server and the OS / applications it is tasked to run. The OS is installed on a SAN LUN and boot from SAN policy is applied to the service profile template or the service profile. If the service profile were to be moved to another server, the pwwn of the HBAs and the Boot from SAN (BFS) policy also moves along with it. The new server now takes the same exact character of the old server, providing the true unique stateless nature of the UCS Blade Server.
The key benefits of booting from the network:
•
•
•
•
•
•
With Boot from SAN, the image resides on a SAN LUN and the server communicates with the SAN through a host bus adapter (HBA). The HBAs BIOS contain the instructions that enable the server to find the boot disk. All of the FC-capable Converged Network Adapter (CNA) cards supported on Cisco UCS B-series blade servers support Boot from SAN.
After power on self-test (POST), the server hardware component fetches the boot device that is designated as the boot device in the hardware BOIS settings. Once the hardware detects the boot device, it follows the regular boot process.
Configuring Boot from SAN Overview
There are three distinct phases during the configuration of Boot from SAN. The high-level procedures are:
1.
2.
3.
In each of the following sections, each high-level phase will be discussed.
SAN Configuration on Nexus 5548UP
The FCoE and NPIV feature has to be turned on in the Nexus 5500 series switch. Make sure you have 8 GB SFP+ modules connected to the Nexus 5548UP ports. The port mode is set to AUTO as well as the speed is set to AUTO. Rate mode is "dedicated" and when everything is configured correctly you should see something like the output below on a Nexus 5500 series switch for a given port (for example, Fc1/17).
Note
Cisco Fabric Manager can also be used to get a overall picture of the SAN configuration and zoning information. As discussed earlier, the SAN zoning is done upfront for all the pwwns of the initiators with the EMC VNX5500 target pwwns.
VDI-N5548-A# show feature | grep npivnpiv 1 enabledVDI-N5548-A# show interface brief-------------------------------------------------------------------------------Interface Vsan Admin Admin Status SFP Oper Oper PortMode Trunk Mode Speed ChannelMode (Gbps)-------------------------------------------------------------------------------fc1/17 1 auto on up swl F 8 --fc1/18 1 auto on up swl F 8 --The FC connection was used for configuring boot from SAN for all of server blades. In addition, a general purpose 1TB infrastructure LUN for infrastructure virtual machine storage and 14 write-cache LUNs for each VDI host were provisioned.
Single vSAN zoning was set up on the Nexus 5548's to make those VNX5500 LUNs visible to the infrastructure and test servers.
An example SAN zone configuration is shown below on the Fabric A side:
VDI-N5548-A# sh zone name B200M3-CH1-SERVER1-FC0 vsan 1zone name B200M3-CH1-SERVER1-FC0 vsan 1member pwwn 20:00:00:25:b5:c1:00:af! [B200M3-CH1-SERVER1-fc0]member pwwn 50:06:01:60:46:e0:5e:0a! [VNX5500-A0]member pwwn 50:06:01:69:46:e0:5e:0a! [VNX5500-B1]VDI-N5548-A# sh zone name B200M3-CH1-SERVER2-FC0 vsan 1zone name B200M3-CH1-SERVER2-FC0 vsan 1member pwwn 20:00:00:25:b5:c1:00:9f! [B200M3-CH1-SERVER2-fc0]member pwwn 50:06:01:60:46:e0:5e:0a! [VNX5500-A0]member pwwn 50:06:01:69:46:e0:5e:0a! [VNX5500-B1]Where 20:00:00:25:b5:c1:00:af /20:00:00:25:b5:c1:00:9f are blade servers pwwn's of their respective Converged Network Adapters (CNAs) that are part of the Fabric A side.
The EMC FC target ports are 50:06:01:60:46:e0:5e:0a /50:06:01:69:46:e0:5e:0a and belong to one port on the FC modules on SP-A and SP-B.
Similar zoning is done on the second Nexus 5548 in the pair to take care of the Fabric B side as shown below:
VDI-N5548-B# sh zone name B200M3-CH1-SERVER1-FC1 vsan 1 zone name B200M3-CH1-SERVER1-FC1 vsan 1member pwwn 20:00:00:25:b5:c1:00:bf[B200M3-CH1-SERVER1-fc1]member pwwn 50:06:01:61:46:e0:5e:0a[VNX5500-A1]member pwwn 50:06:01:68:46:e0:5e:0a[VNX5500-B0]VDI-N5548-B# sh zone name B200M3-CH1-SERVER2-FC1 vsan 1zone name B200M3-CH1-SERVER2-FC1 vsan 1member pwwn 20:00:00:25:b5:c1:00:8f[B200M3-CH1-SERVER2-fc1]member pwwn 50:06:01:61:46:e0:5e:0a[VNX5500-A1]member pwwn 50:06:01:68:46:e0:5e:0a[VNX5500-B0]Where 20:00:00:25:b5:c1:00:bf /20:00:00:25:b5:c1:00:8f are blade servers pwwn's of their respective Converged Network Adapters (CNAs) that are part of the Fabric B side.
The EMC FC target ports are 50:06:01:61:46:e0:5e:0a /50:06:01:68:46:e0:5e:0a and belong to the other port on the FC modules on SP-A and SP-B. They were spread across the two controllers for redundancy as shown in Figure 16.
Figure 16 VNX5500 FC Target Ports
For detailed Nexus 5500 series switch configuration, refer to Cisco Nexus 5500 Series NX-OS SAN Switching Configuration Guide. (See the Reference Section of this document for the link.)
Configuring Boot from SAN on EMC VNX
The steps required to configure boot from SAN LUNs on EMC VNX are as follows:
1.
2.
3.
4.
5.
6.
When the Cisco UCS Blade Server boots up, its vHBAs will connect to the provisioned EMC Boot LUNs and the hypervisor operating system can be installed.
SAN Configuration on Cisco UCS Manager
To enable Boot from SAN on the Cisco UCS Manager 2.1 (UCS-M) series, do the following:
1.
2.
3.
4.
VDI-N5548-A# show fcns database vsan 10xe300ef N 50:06:01:60:46:e0:5e:0a (Clariion) scsi-fcp:both0xe301ef N 50:06:01:69:46:e0:5e:0a (Clariion) scsi-fcp:bothVDI-N5548-B # show fcns database vsan 10x470400 N 50:06:01:61:46:e0:5e:0a (Clariion) scsi-fcp0x470500 N 50:06:01:68:46:e0:5e:0a (Clariion) scsi-fcp
5.
6.
7.
At the end your Boot from SAN policy will look like the following:
8.
EMC VNX5500 Storage Configuration
Figure 17 illustrates the layout of the disks that are required to store 2,000 desktop virtual machines. This layout does not include space for user profile data. Refer to "VNX shared file systems" on page 23 section for more information.
Figure 17 Core Storage Layout
Core Storage Layout Overview
The following core configuration is used in the reference architecture:
•
•
•
•
•
•
•
•
•
In solution validation testing, storage space for user data and infrastructure was allocated on the VNX array as shown in the following figure. This storage is in addition to the core storage shown above. If storage for user data exists elsewhere in the production environment, this storage is not required.
EMC Storage Configuration for VMware View
EMC Storage Configuration for VMware ESXi 5.1 Infrastructure Clusters
The steps required to configure LUNs for the VDI Datastores are as follows:
1.
2.
Two LUNs of 2.08TB are carved out of the RAID 5 storage pool configured with 10 SAS drives. The LUNs are used to store infrastructure virtual machines such as VMware Horizon View 5.2 Connection Servers, View Composer servers, and VMware vCenter server.
Example EMC Boot LUN Configuration
Each ESXi server requires a boot LUN from SAN for the hypervisor OS. A total of 43 LUNs are carved out of the 5-disk RAID 5 pool. Each LUN is 5GB in size.
EMC FAST Cache in Practice
EMC FAST Cache uses Flash drives to add an extra layer of cache between the dynamic random access memory (DRAM) cache and rotating disk drives, thereby creating a faster medium for storing frequently accessed data. FAST Cache is an extendable Read/Write cache. It boosts application performance by ensuring that the most active data is served from high-performing Flash drives and can reside on this faster medium for as long as is needed.
FAST Cache tracks data activity at a granularity of 64KB and promotes hot data in to FAST Cache by copying from the hard disk drives (HDDs) to the Flash drives assigned to FAST Cache. Subsequent IO access to that data is handled by the Flash drives and is serviced at Flash drive response times-this ensures very low latency for the data. As data ages and becomes less active, it is flushed from FAST Cache to be replaced by more active data.
Only a small number of Flash drives are needed enabling FAST Cache to provide greater performance increases than implementing a large number of short-stroked HDDs. This results in cost savings in data center space, power, and cooling requirements that lowers overall TCO for the business.
FAST Cache is particularly suited to applications that randomly access storage with high frequency, such as Oracle and SQL OLTP databases. OLTP databases have inherent locality of reference with varied IO.
Cisco UCS Manager Configuration for VMware ESXi 5.1
This section addresses creation of the service profiles and VLANs to support the project.
Service Profile Templates
Two types of service profiles were required to support two different blade server types (Table 7).
Table 7 Role/Server/OS Deployment
To support those different hardware platforms, service profile templates were created, utilizing various policies created earlier.
The service profile templates were then used to quickly deploy service profiles for each blade server in the Cisco UCS system. When each blade server booted for the first time, the service profile was deployed automatically, providing the perfect configuration for the VMware ESXi 5.1 installation.
VLAN Configuration
In addition, to control network traffic in the infrastructure and assure priority to high value traffic, virtual LANs (VLANs) were created on the Nexus 5548s, on the Cisco UCS Manager (Fabric Interconnects,) and on the Nexus 1000V Virtual Switch Modules in each vCenter Cluster. The virtual machines in the environment used the VLANs depending on their role in the system.
A total of seven Virtual LANs, VLANs, were utilized for the project. Table 8 identifies them and describes their use.
Table 8 VLAN Naming and Use
VLANs are configured using Cisco UCS Manager from the LAN tab; LAN\VLANs node in the left pane of Cisco UCS Manager. They were set up earlier in section 7.2.1 Base Cisco UCS System Configuration.
Installing and Configuring ESXi 5.1
In this study, we used Fibre Channel storage to boot the hosts from LUNs on the VNX5500 storage system. Prior to installing the operating system, storage groups are created, assigning specific boot LUNs to individual hosts. (See section Configuring Boot from SAN on EMC VNX for details.)
VMware ESXi 5.1 can be installed in boot-from-SAN mode using standard hypervisor deployment techniques including:
•
•
Install VMware ESXi 5.1
ESXi was installed using the UCS Manager KVM console with the Cisco Customized ESXi 5.1 ISO image mounted. The Cisco UCS Manager boot policy deployed to each blade was set to boot from CD then SAN to accommodate hypervisor installs or updates.
The IP address, hostname, and NTP server were configured using Direct Console ESXi Interface accessed from Cisco UCS Manager KVM console; http://pubs.vmware.com/vsphere-51/index.jsp?topic=%2Fcom.vmware.vsphere.install.doc_50%2FGUID-26F3BC88-DAD8-43E7-9EA0-160054954507.html
Install and Configure vCenter
To manage hypervisors and virtual machines a dedicated vCenter server instance was installed on Windows 2008R2 SP1 based virtual machine.
To support vCenter instance a Microsoft SQL Server 2008 R2 server was created to host vCenter database. Refer to the Microsoft documentation about configuring SQL Server and SQL Server clusters: http://msdn.microsoft.com/en-us/library/ms189134(v=sql.105).aspx and http://msdn.microsoft.com/en-us/library/ms189134(v=sql.105).aspx)
Install and Configure vCenter
1.
2.
3.
4.
5.
6.
Install VMware Licenses
1.
2.
3.
ESXi 5.1 Cluster Configuration
To accommodate maximum recommendations for View 5.2 on ESXi 5.1 we created four ESXi 5.1 clusters described below. For each of the two VDI clusters using seven Cisco UCS B200 M3 Blade Servers and 1000 virtual machines being hosted.
The 14 Cisco UCS B200 M3s and six Cisco UCS B250 M2 ESX hosts were configured into four Clusters:
•
•
•
•
Infra-Cluster Infrastructure Cluster
The Infra-Cluster cluster was used to host all of the virtualized servers within the VDA Infrastructure, including two pairs of Nexus 1000V Virtual Switch Manager (VSM) appliances, one for each virtual desktop cluster.
Two physical Cisco UCS B250-M2 hosts were used in this cluster.
One standard switch to manage VMware Management, VDA, vMotion, and Storage traffic were configured on DC-INF cluster hosts. Three pairs of fault tolerant VSMs introduced the N1KV Management, Control and Packet VLANs to the environment.
Virtual Desktop Clusters
The following clusters were used to host 2000 desktops:
VDI-CL1-N1KVVDI-CL2-N1KV
Each of these desktop clusters were configured identically with a Nexus 1000V high availability distributed virtual switch providing the required network connectivity.
The Nexus 1000V switches were configured to manage networking for all three ESX Clusters hosting virtual desktops, working in concert with the Cisco UCS Fabric Interconnects and Nexus 5548UP access layer switches to provide end to end Quality of Service for network communications, insuring the highest quality virtual desktop end user experience.
The Nexus 1000V configuration is described in detail in section Nexus 1000V Configuration in L3 Mode.
Login VSI Launcher Cluster
The separate Launcher-CL cluster was used to host Login Consultants' LoginVSI launcher VMs and a LoginVSI console VM. It was hosted on the same Cisco UCS Domain with dedicated storage.
The Laucher-CL cluster utilized the Standard vSwitch configured as shown below:
Installing and configuring VMware View 5.2
Building out a VMware View 5.2 environment requires the installation of the following components:
•
•
•
•
•
This section outlines the tasks required to build the View 5.2 environment used in this study. Refer to the VMware View Installation guide for View 5.2 for more details:
https://pubs.vmware.com/view-52/topic/com.vmware.ICbase/PDF/horizon-view-52-installation.pdf.
Pre-requisites
The following is a list of pre-requisites that are required with installing View 5.2 components. They are as follows:
•
–
–
Note
•
•
•
•
–
–
–
–
Create SQL Databases for View 5.2
View Manager Installer needs a separate database for View Composer Server and View Server events.
Create Database for View Composer Server
1.
2.
3.
Create Event Database for View Administrator.
Create a Database for View Administrator Events and user with SA authentication.
Install View Manager and components
Download View Manager software from the link given below.
https://my.vmware.com/web/vmware/details?downloadGroup=VIEW-520-PREMIER&productId=320&rPId=3908
Install View Connection server
Log into View Connection server with Domain Administrator credentials.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Note
12.
Install View Replica Server
1.
2.
3.
4.
5.
6.
7.
8.
9.
Install View Composer Server
View Composer server can be install on a separate stand alone server or on the same server that was used for vCenter server installer. For our test, we installed View Composer server on the same server we used for vCenter server.
1.
2.
3.
4.
5.
6.
7.
8.
View Administrator Configuration
To configure the View 5.2 system, follow these steps:
1.
2.
3.
4.
5.
6.
For this test case we used the following parameters.
Max concurrent vCenter Provisioning operations: 20
Max concurrent Power operations: 10
Max concurrent View Composer maintenance operations: 50
Max concurrent View Composer provisioning operations: 50
7.
8.
Note
9.
10.
11.
12.
Note
13.
14.
15.
16.
Install SSL Certificate for View Connection and Replica Server
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
For more information about VMare's best practices to obtain the self-signed certificate, go to: http://pubs.vmware.com/view-52/topic/com.vmware.ICbase/PDF/horizon-view-52-obtaining-certificates.pdf
Install View Client on End Points
1.
https://my.vmware.com/web/vmware/info/slug/desktop_end_user_computing/vmware_horizon_view/5_2?rct=j&q=cmware%20view%205.2%20horizon%20download&source=web&cd=1&ved=0CEIQFjAA&url=http://www.vmware.com/go/downloadview
2.
3.
4.
5.
6.
7.
8.
9.
Configure the View 5.2 Hosts and Storage
Configure Content Based Read Cache (CBRC) on View 5.2 Hosts
CBRC was introduced as a feature of vSphere 5. It is a read cache that is particularly useful during boot storms. It becomes an essential configuration for floating assignment View 5.2 Linked Clones.
The CBRC feature provides a per-host RAM-based solution for View desktops. This considerably reduces the read I/O requests that are issued to the storage layer, and also addresses boot storm snags.
CBRC is configured in vCenter by highlighting the host; access the Configuration Tab, Software, and Advanced Settings.
Each ESXi host used for View Desktops we enabled CBRC and increased the CBRC.DCacheMemReserved to 2048.
These CBRC settings are used in conjunction with the View 5.2 Administrator, View Configuration, Servers, vCenter Server Properties, Host Cashing tab.
In our test environment, we enabled 2GB of CBRD and correspondingly, 2GB of Host Cache in View Administrator. This combination enables the View Storage Accelerator feature.
Storage Configuration for View 5.2 Hosts
On VNX 5500 30 SAS disks with 300 GB capacity were used to create 16 LUNs, each with a capacity of 370 GB. 2 LUNs with capacities of 50 GB each were created to store replica disks.
Each ESXi host in each cluster was assigned 8 LUNs as VMFS5 datastores for linked clones and one 50Gb VMFS5 datastore to hold the Replica disk.
Note
Desktop Delivery Base Image Creation and Desktop Deployment
Microsoft Windows 7 Golden Image Creation
Create base Windows SP1 Virtual Machine
1.
For the virtual machine following parameters were used:
Memory : 1536Mb
Processor : 1vCPU
Hard Disk : 18Gb
Network Adapter : 1 attached to VDI port-group on Nexus 1000v
2.
3.
4.
a.
b.
c.
5.
6.
This complete the process of creating the Golden Image virtual machine.
Optimization of Base Windows 7 SP1 Virtual Machine
1.
www.vmware.com/files/pdf/VMware-View-OptimizationGuideWindows7-EN.pdf
Install View 5.2 Virtual Desktop Agent Software
1.
https://my.vmware.com/web/vmware/details?downloadGroup=VIEW-520-PREMIER&productId=320&rPId=4175
2.
3.
4.
5.
6.
Install Additional Software
1.
•
•
2.
3.
4.
Perform Additional View 5.2 Configuration
Create a Snapshot for the Virtual Machine
1.
2.
3.
4.
Create Customization Specification for Virtual Desktops
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
Configure the View Desktop Pools and Options.
Desktop Pools are the containment object in View 5.2 Administrator that hold the configuration and the provisioned linked clones in the View environment.
The maximum recommended number of virtual machines in a VMware ESXi cluster is 1000. Therefore, we created two View 5.2 pools with identical settings to match up with our two ESXi 5.1 clusters for VDI described earlier in this document.
The following sections describe how we configured our View 5.2 environment.
Create the Desktop Pools
1.
There are three types of Desktop Pools you can create and descriptions for each type is given on the right side of the screen.
2.
There are two types of User Assignment options available:
•
•
Note
3.
4.
Note
5.
6.
7.
•
•
Note
•
Note
8.
9.
10.
Note
11.
12.
13.
14.
15.
16.
17.
After the completion of the pool setting, a replica from the parent VM is created and the provisioning of the desktops as per the pool settings begins.
Test Setup and Configurations
In this project, we tested a single Cisco UCS B200 M3 blade in a single chassis and fourteen Cisco UCS B200 M3 blades in two chassis to illustrate linear scalability.
Cisco UCS Test Configuration for Single Blade Scalability
Figure 18 illustrates the Cisco UCS test configuration single-blade scalability.
Figure 18 Cisco UCS B200 M3 Blade Server for Single Server Scalability
Hardware components
•
•
•
•
•
•
•
•
Software components
•
•
•
•
•
Cisco UCS Configuration for Two Chassis—Fourteen Blade Test
Figure 19 illustrates the two chassis test configuration.
Figure 19 Two Chassis Test Configuration-14 x B200 M3 Blade Servers
Hardware components
•
•
•
•
•
•
•
•
Software components
•
•
•
•
•
Testing Methodology and Success Criteria
All validation testing was conducted on-site within the Cisco Labs with joint support from both Cisco and EMC resources.
The testing results focused on the entire process of the virtual desktop lifecycle by capturing metrics during the desktop boot-up, user logon and virtual desktop acquisition (also referred to as ramp-up,) user workload execution (also referred to as steady state), and user logoff for the Hosted VDI model under test.
Test metrics were gathered from the hypervisor, virtual desktop, storage, and load generation software to assess the overall success of an individual test cycle. Each test cycle was not considered passing unless all of the planned test users completed the ramp-up and steady state phases (described below) and unless all metrics were within the permissible thresholds as noted as success criteria.
Three successfully completed test cycles were conducted for each hardware configuration and results were found to be relatively consistent from one test to the next.
Load Generation
Within each test environment, load generators were utilized to put demand on the system to simulate multiple users accessing the Horizon View 5.2 environment and executing a typical end-user workflow. To generate load within the environment, an auxiliary software application was required to generate the end user connection to the Horizon View environment, to provide unique user credentials, to initiate the workload, and to evaluate the end user experience.
In the Hosted VDI test environment, sessions launchers were used simulate multiple users making a direct connection to Horizon View 5.2 via a VMware Horizon View PCoIP protocol connection.
User Workload Simulation—LoginVSI From Login Consultants
One of the most critical factors of validating a Horizon View deployment is identifying a real-world user workload that is easy for customers to replicate and standardized across platforms to allow customers to realistically test the impact of a variety of worker tasks. To accurately represent a real-world user workload, a third-party tool from Login Consultants was used throughout the Hosted VDI testing.
The tool has the benefit of taking measurements of the in-session response time, providing an objective way to measure the expected user experience for individual desktop throughout large scale testing, including login storms.
The Virtual Session Indexer (Login Consultants' Login VSI 3.6) methodology, designed for benchmarking Server Based Computing (SBC) and Virtual Desktop Infrastructure (VDI) environments is completely platform and protocol independent and allows customers to easily replicate the testing results in their environment.
Note
•
•
Like real users, the scripted Login VSI session will leave multiple applications open at the same time. The medium workload is the default workload in Login VSI and was used for this testing. This workload emulated a medium knowledge working using Office, IE, printing and PDF viewing.
•
•
•
•
•
Each loop will open and use:
•
•
•
•
•
•
•
•
A graphical representation of the medium workload is shown below.
You can obtain additional information on Login VSI from http://www.loginvsi.com.
Testing Procedure
The following protocol was used for each test cycle in this study to help ensure consistent results.
Pre-Test Setup for Single and Multi-Blade Testing
•
•
•
Test Run Protocol
To simulate severe, real-world environments, Cisco requires the log-on and start-work sequence, known as Ramp Up, to complete in 30 minutes. Additionally, we require all sessions started, whether 155 single server users or 2000 full scale test users to become active within 2 minutes after the session is launched.
For each of the three consecutive runs on single blade (155 User) and 14-blade (2000 User) tests, the same process was followed:
1.
–
–
–
–
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
Success Criteria
There were multiple metrics that were captured during each test run, but the success criteria for considering a single test run as pass or fail was based on the key metric, VSI Max. The Login VSI Max evaluates the user response time during increasing user load and assesses the successful start-to-finish execution of all the initiated virtual desktop sessions.
Login VSI Max
VSI Max represents the maximum number of users the environment can handle before serious performance degradation occurs. VSI Max is calculated based on the response times of individual users as indicated during the workload execution. The user response time has a threshold of 4000ms and all users response times are expected to be less than 4000ms in order to assume that the user interaction with the virtual desktop is at a functional level. VSI Max is reached when the response times reaches or exceeds 4000ms for 6 consecutive occurrences. If VSI Max is reached, that indicates the point at which the user experience has significantly degraded. The response time is generally an indicator of the host CPU resources, but this specific method of analyzing the user experience provides an objective method of comparison that can be aligned to host CPU performance.
Note
Calculating VSIMax
Typically the desktop workload is scripted in a 12-14 minute loop when a simulated Login VSI user is logged on. After the loop is finished it will restart automatically. Within each loop the response times of seven specific operations is measured in a regular interval: six times in within each loop. The response times if these seven operations are used to establish VSImax.
The seven operations from which the response times are measured are:
•
–
•
–
•
–
•
–
•
–
•
–
•
–
These measured operations with Login VSI do hit considerably different subsystems such as CPU (user and kernel), Memory, Disk, the OS in general, the application itself, print, GDI, etc. These operations are specifically short by nature. When such operations are consistently long: the system is saturated because of excessive queuing on any kind of resource. As a result, the average response times will then escalate. This effect is clearly visible to end-users. When such operations consistently consume multiple seconds the user will regard the system as slow and unresponsive.
With Login VSI 3.0 and later it is now possible to choose between 'VSImax Classic' and 'VSImax Dynamic' results analysis. For these tests, we utilized VSImax Dynamic analysis.
VSIMax Dynamic
VSImax Dynamic is calculated when the response times are consistently above a certain threshold. However, this threshold is now dynamically calculated on the baseline response time of the test.
Five individual measurements are weighted to better support this approach:
•
•
•
•
•
•
•
A sample of the VSImax Dynamic response time calculation is displayed below:
Then the average VSImax response time is calculated based on the amount of active Login VSI users logged on to the system. For this the average VSImax response times need to consistently higher than a dynamically calculated threshold.
To determine this dynamic threshold, first the average baseline response time is calculated. This is done by averaging the baseline response time of the first 15 Login VSI users on the system.
The formula for the dynamic threshold is: Avg. Baseline Response Time x 125% + 3000. As a result, when the baseline response time is 1800, the VSImax threshold will now be 1800 x 125% + 3000 = 5250ms.
Especially when application virtualization is used, the baseline response time can wildly vary per vendor and streaming strategy. Therefore it is recommend to use VSImax Dynamic when comparisons are made with application virtualization or anti-virus agents. The resulting VSImax Dynamic scores are aligned again with saturation on a CPU, Memory or Disk level, also when the baseline response time are relatively high.
Determining VSIMax
The Login VSI analyzer will automatically identify the "VSImax". In the example below the VSImax is 99. The analyzer will automatically determine "stuck sessions" and correct the final VSImax score.
•
•
Figure 20 Sample Login VSI Analyzer Graphic Output
•
•
•
•
Note
Our Login VSI "pass" criteria, accepted by Login Consultants for this testing is as follows:
•
•
•
•
–
–
•
•
The purpose of this testing is to provide the data needed to validate VMware View 5.2 automated pool, floating assignment linked clone virtual desktops using ESXi 5.1 and vCenter 5.1 to virtualize Microsoft Windows 7 SP1 desktops on Cisco UCS B200 M3 blade servers using a EMC VNX5500 storage system.
The information contained in this section provides data points that a customer may reference in designing their own implementations. These validation results are an example of what is possible under the specific environment conditions outlined here, and do not represent the full characterization of View 5.2 with VMware vSphere.
Two test sequences, each containing three consecutive test runs generating the same result, were performed to establish single server performance and multi-server, linear scalability.
One additional series of stress tests on a single blade server was conducted to establish the official Login VSI Max Score. To reach the Login VSI Max, we ran 195 Medium Workload (with flash) Windows 7 SP1 sessions on a single server. The Login VSI score was achieved on three consecutive runs and is shown in the next section of the document.
Cisco UCS Test Configuration for Single-Server Scalability Test Results
This section details the results from the View 5.2 Hosted VDI single blade server validation testing. The primary success criteria used to validate the overall success of the test cycle is an output chart from Login Consultants' VSI Analyzer Professional Edition, VSIMax Dynamic for the Medium workload (with Flash.)
Note
We ran the single server test at approximately 10% lower user density than prescribed by the Login VSI Max to achieve a successful pass of the test with server hardware performance in a realistic range.
Additionally, graphs detailing the CPU, Memory utilization and network throughput during peak session load are also presented. Given adequate storage capability, the CPU utilization determined the maximum VM density per blade.
The charts below present our recommended maximum Login VSI Medium workload loading on a single blade server.
Figure 21 Login VSI Max graph- 155 Sessions Average Response Time below 2000 m seconds
The following graphs detail CPU, Memory, Disk and Network performance on the Single Cisco UCS B200-M3 Blades
Figure 22 155 User Single B200 M3 CPU Utilization Boot Phase
Figure 23 155 User Single B200 M3 Available Memory Boot Phase
Figure 24 155 User Single B200 M3 Cisco VIC1240 MLOM Mbps Received/TransmittedBoot Phase
Figure 25 155 User Single B200 M3 Cisco VIC1240 MLOM Physical Disk Adapter MBps Read/Write Boot Phase
Figure 26 155 User Single B200 M3 CPU Core Utilization - Test Phase
Figure 27 155 User Single B200 M3 CPU Processor Total Utilization Time - Test Phase
Figure 28 155 User Single B200 M3 Available Memory Test Phase
Figure 29 155 User Single B200 M3 Cisco VIC1240 MLOM Mbps Received/Transmitted Test Phase
Figure 30 155 User Single B200 M3 Cisco VIC1240 MLOM Mbps Read/Written Test Phase
The following charts details for VIEW Connection Server performance during the single blade, 155 User test: Boot Phase
Figure 31 155 User View Connection Server 5.2 CPU Utilization
Figure 32 155 User View Connection Server 5.2 Available Memory
Figure 33 155 User View Connection Server 5.2 Bytes Received /Second
Figure 34 155 User View Connection Server 5.2 Bytes Sent/Second
Figure 35 155 User View 5.2 - View Connection Server CPU Utilization-Test Phase
Figure 36 155 User View 5.2 - View Connection Total User Time- TEST Phase
Figure 37 155 User View 5.2 - View Connection Server Available Memory
Figure 38 155 User View 5.2 View Connection Server Bytes Received
Figure 39 155 User View 5.2 View Connection Server Bytes Sent
Cisco UCS Test Configuration for 2000 Desktop Two-Cluster Scalability Test Results
This section details the results from the View 5.2 Hosted VDI seven blade server, two-cluster, 2000 user validation testing. It demonstrates linear scalability for the system. The primary success criteria used to validate the overall success of the test cycle is an output chart from Login Consultants' VSI Analyzer Professional Edition, VSIMax Dynamic for the Medium workload (with Flash.)
Note
We ran the multi-server test at an average user density slightly below 143 users per blade across the system. Two ESX Clusters, each containing seven B200 M3s ran the entire workload. In fact the fourteen blade test harness provides N+1 server fault tolerance on a system basis to achieve a successful pass of the test with server hardware performance in a realistic range.
Additionally, graphs detailing the CPU, Memory utilization and network throughput during peak session load are also presented for a representative blade running 143 user sessions below. We have provided performance charts for all 14 blades in Appendix B to illustrate this point.
Given adequate storage capability, the blade CPU utilization determined the maximum recommended VM density per blade for the 2000 user environment.
We also present performance information on key infrastructure virtual machines with the tested blade data.
For the large scale test, we are including the EMC VNX5500 performance metrics as well.
Figure 40 2000 Desktop Sessions on VMware ESXi 5.1 below 4000 ms
The following graphs detail CPU, Memory, Disk and Network performance on a representative Cisco UCS B200 M3 Blade during the fourteen blade, 2000 User test. (Representative results for all fourteen blades in one of the vCenter clusters can be found in Appendix B.)
Figure 41 2000 User Single B200 M3 Core CPU Utilization Boot Phase
Figure 42 2000 User Single B200 M3 CPU Utilization Boot Phase
Figure 43 2000 User Single Cisco UCS B200 M3 Cisco NonKernel Mbytes Available Boot Phase
Figure 44 2000 User Single Cisco B200 M3 Cisco VIC1240 MLOM Physical Disk Adapter
Figure 45 2000 User Single Cisco UCS B200 M3 Cisco VIC1240 MLOM VIC Mbps Receive/Transmit Boot Phase
Figure 46 2000 User Single Cisco UCS B200 M3 CPU Utilization Test Phase
Figure 47 2000 User Single Cisco UCS B200 M3 Cisco VIC1240 MLOM Network Adapter Mbps Receive/Transmit Test Phase
Figure 48 2000 User Single Cisco UCS B200 M3 Cisco VIC1240 MLOM Physical Disk Adapter MBps Read/Write Test Phase
Figure 49 2000 User Single Cisco UCS B200 M3 Cisco VIC1240 MLOM VIC Mbps Receive/Transmit Test Phase
Figure 50 2000 User Single Cisco UCS B200 M3 Cisco Non-Kernel Mbytes Avaaible
The following charts detail the VNX5500 performance during the fourteen blade, 2000 User test:
Figure 51 2000 Users EMC VNX5500 SP Utilization Boot Phase
Figure 52 2000 Users EMC VNX5500 SP Queue Lengths Boot Phase
Figure 53 2000 Users EMC VNX5500 SP Total Throughput Boot Phase
Figure 54 2000 Users EMC VNX5500 SP Utilization Test Phase
Figure 55 2000 Users EMC VNX5500 SP Queue Lengths Test Phase
Figure 56 2000 Users EMC VNX5500 SP Total Throughput Test Phase
Figure 57 2000 Users EMC VNX5500 SP Total Response Time Test Phase
The following charts detail infrastructure server performance during the fourteen blades, 2000 User test:
Figure 58 2000 User View Connection Server 5.2 CPU Utilization Test Phase
Figure 59 2000 Users EMC VNX5500 SP Queue Lengths Boot Phase
Figure 60 2000 Users EMC VNX5500 SP Total Throughput Boot Phase
Figure 61 2000 Users EMC VNX5500 SP Utilization Test Phase
Figure 62 2000 Users EMC VNX5500 SP Queue Lengths Test Phase
Figure 63 2000 Users EMC VNX5500 SP Total Throughput Test Phase
Figure 64 2000 Users EMC VNX5500 SP Total Response Time Test Phase
The following charts detail infrastructure server performance during the fourteen blades, 2000 User test:
Figure 65 2000 User View Connection Server 5.2 CPU Utilization Test Phase
Figure 66 2000 User View Connection Server 5.2 Available Memory Test Phase
Figure 67 2000 User View Connection Server 5.2 Bytes Received/Second Test Phase
Figure 68 2000 User View Connection Server 5.2 Bytes Sent/Second Test Phase
Scalability Considerations and Guidelines
There are many factors to consider when you begin to scale beyond 2000 User, six chassis, 14 VDI host server configuration, which this reference architecture has successfully tested. In this section we give guidance to scale beyond the 2000 user system.
Cisco UCS System Configuration Considerations
As our results indicate, we have proven linear scalability in the Cisco UCS Reference Architecture as tested.
•
•
•
•
Of course, the backend storage has to be scaled accordingly, based on the IOPS considerations as described in the EMC scaling section. Please refer the EMC section that follows this one for scalability guidelines.
VMware View 5.2 Considerations
VMware View Composer can create and provision up to 1000 desktops per pool when deployed on vSphere 4.1 or later. View Composer can also perform a recompose operation on up to 1,000 desktops at a time. Desktop pool size is limited by the following factors:
•
•
•
A single VMware View Connection server can host up to 2000 simultaneous connections over any supported connection type. Seven View Connection Servers (5 active plus 2 spares) can host up to 10000 direct, RDP or PCoIP connections simultaneously. The sever View Connection Server cluster configuration should not be clustered across WAN links.
VMware View deployments can use VMware HA clusters to guard against physical server failures. With View 5.2 and later and vSphere 5 and later, if you use View Composer and store replica disks on NFS datastores, the cluster can contain up to 32 servers, or nodes.
With vCenter 4.1 and 5.0, each vCenter Server can support up to 10,000 virtual machines.
For more information on VMware View 5.2 configuration and guidelines, see Chapter 11 References.
EMC VNX Storage Guidelines for Horizon View 5.2 Provisioned Virtual Machines
Sizing VNX storage system to meet virtual desktop IOPS requirement is a complicated process. When an I/O reaches the VNX storage, it is served by several components such as Data Mover (NFS), backend dynamic random access memory (DRAM) cache, FAST Cache, and disks. To reduce the complexity, EMC recommends using a building block approach to scale to thousands of virtual desktops.
For more information on storage sizing guidelines to implement virtual desktop infrastructure in VNX unified storage systems, refer to the EMC white paper "Sizing EMC VNX Series for VDI workload - An Architectural Guideline".
VMware ESXi 5.1 Guidelines for Virtual Desktop Infrastructure
In our test environment two adjustments were performed to support our scale:
•
•
For further explanations on a basis for these adjustments and details on how to perform them, refer to the VMware documentation sited in the References section of this document.
References
This section provides links to additional information for each partner's solution component of this document.
Cisco Reference Documents
Third-Generation Fabric Computing: The Power of Unification webcast replay
Cisco Unified Computing System Manager Home Page
http://www.cisco.com/en/US/products/ps10281/index.html
Cisco UCS B200 M3 Blade Server Resources
http://www.cisco.com/en/US/partner/products/ps12288/index.html
Cisco UCS 6200 Series Fabric Interconnects
http://www.cisco.com/en/US/partner/products/ps11544/index.html
Cisco Nexus 1000V Series Switches Resources
http://www.cisco.com/en/US/partner/products/ps9902/index.html
Cisco Nexus 5500 Series Switches Resources
http://www.cisco.com/en/US/products/ps9670/index.html
Download Driver Software for Cisco UCS B200 M3 Blade Server
Download Cisco UCS Manager and Blade Software Version 2.1(1b)
Download Cisco UCS Central Software Version 1.0(1a)
http://software.cisco.com/download/cart.html?imageGuId=8CAAAD77B3A1DB35B157BE84ED109A4703849F53&i=rs
VMware View Reference Documents
View 5 Documents
Performance and Best Practices
http://www.vmware.com/files/pdf/view/vmware-horizon-view-best-practices-performance-study.pdf
View 5.2 Architecture and Planning https://pubs.vmware.com/view-52/topic/com.vmware.ICbase/PDF/horizon-view-52-architecture-planning.pdf
View 5 with PCoIP Network Optimization Guide
http://www.vmware.com/files/pdf/view/VMware-View-5-PCoIP-Network-Optimization-Guide.pdf
Virtual Desktop
Windows 7 Optimization Guide http://www.vmware.com/files/pdf/VMware-View-OptimizationGuideWindows7-EN.pdf
EMC Reference Documents
•
VMware Reference Documents
•
•
•
•
•
Appendix
Click the following link to open the Appendix: http://www.cisco.com/en/US/docs/unified_computing/ucs/UCS_CVDs/ucs_vspex_vview5.2_2000_appendix.pdf.
Feedback