EMC VSPEX Solution for Infrastructure as a Service with VMware

DESIGN AND IMPLEMENTATION GUIDE

EMC VSPEX SOLUTION FOR INFRASTRUCTURE AS A SERVICE WITH VMWARE VCLOUD SUITE

EMC VSPEX

Abstract

This Design and Implementation Guide describes how to design virtualized VMware vCloud Suite resources on EMC® VSPEX® Private Cloud solutions. It also illustrates how to size vCloud Suite components using VSPEX guidelines, allocate resources following best practices, and use all the benefits that VSPEX offers.

January 2014

2 EMC VSPEX Solution for Infrastructure as a Service with VMware vCloud Suite Design and Implementation Guide

Copyright © 2014 EMC Corporation. All rights reserved. Published in the USA.

Published January 2014

EMC believes the information in this publication is accurate as of its publication date. The information is subject to change without notice.

The information in this publication is provided as is. EMC Corporation makes no representations or warranties of any kind with respect to the information in this publication, and specifically disclaims implied warranties of merchantability or fitness for a particular purpose. Use, copying, and distribution of any EMC software described in this publication requires an applicable software license.

EMC2, EMC, and the EMC logo are registered trademarks or trademarks of EMC Corporation in the United States and other countries. All other trademarks used herein are the property of their respective owners.

For the most up-to-date listing of EMC product names, see EMC Corporation Trademarks on EMC.com.

EMC VSPEX Solution for Infrastructure as a Service with VMware vCloud Suite Design and Implementation Guide

Part Number H12432

http://www.emc.com/legal/emc-corporation-trademarks.htm

http://www.emc.com/legal/emc-corporation-trademarks.htm

Contents


Contents

Chapter 1 Introduction 9

Introduction ............................................................................................................. 10

Overview .............................................................................................................. 10

Essential reading ................................................................................................. 10

Purpose of this guide ................................................................................................ 11

Business value ......................................................................................................... 11

Scope ....................................................................................................................... 12

Audience .................................................................................................................. 12

Terminology.............................................................................................................. 12

Chapter 2 Solution Overview 13

Overview .................................................................................................................. 14

Key components ....................................................................................................... 14

VMware vCloud Suite ........................................................................................... 15

EMC VNX .............................................................................................................. 16

EMC Integration for vCloud Suite .......................................................................... 16

Hardware list ....................................................................................................... 17

Software list ........................................................................................................ 17

Chapter 3 Solution Design 19

Overview .................................................................................................................. 20

Solution architecture ................................................................................................ 20

High-level solution architecture ........................................................................... 20

Storage design ......................................................................................................... 21

VNX storage tiering and FAST VP .......................................................................... 22

Storage mapping ................................................................................................. 23

vCloud Suite design .................................................................................................. 24

VMware vSphere .................................................................................................. 24

vCloud Automation Center ................................................................................... 26

vCenter Operations Manager ............................................................................... 27

vCloud Networking and Security .......................................................................... 28

Design considerations and best practices ................................................................ 30

Network design .................................................................................................... 30

Storage design..................................................................................................... 30

Virtualization design ................................................................................................ 31

Contents


Chapter 4 Solution Implementation 33

Overview .................................................................................................................. 34

Installing and configuring vCloud Management cluster ............................................. 34

Setting up and configuring SQL Server database .................................................. 34

Deploying management and production resource cluster vCenter servers ............ 35

Preparing a management cluster.......................................................................... 35

EMC integration ................................................................................................... 35

Deploying vCloud Networking and Security ............................................................... 47

Deploying vCAC ........................................................................................................ 48

Database server ................................................................................................... 48

vCAC web server .................................................................................................. 48

vCAC server Manager Service ............................................................................... 48

DEM worker server ............................................................................................... 49

vCloud Suite integration (configuring endpoint) ................................................... 49

Deploying vCenter Operations Manager .................................................................... 50

vCloud Suite Integration ...................................................................................... 50

EMC Integration ................................................................................................... 50

Chapter 5 Solution Verification 52

Overview .................................................................................................................. 53

Verifying vSphere implementation ............................................................................ 53

Preparing production vCenter resources .............................................................. 53

VSI operations ..................................................................................................... 53

Configure datastore clusters in vSphere ............................................................... 56

Verifying vCNS implementation ................................................................................ 57

Separating tenant networks ................................................................................. 57

Verifying vCAC implementation ................................................................................. 58

Cloud roles and resource administration ............................................................. 58

Creating vSphere virtual resource reservation and cost profile ............................. 61

vCAC blueprints ................................................................................................... 64

Deploy virtual machines from blueprints .............................................................. 68

Reconfigure existing virtual machine ................................................................... 68

Decommission a virtual machine ......................................................................... 70

Chargeback and metering .................................................................................... 70

vCenter Operations Management ............................................................................. 78

Monitoring vCenter objects and resources ........................................................... 78

Monitoring VNX health and resources .................................................................. 82

Capacity planning and optimization ..................................................................... 85

Contents


Chapter 6 Resource Sizing Guide 90

Overview .................................................................................................................. 91

Cloud management environment sizing .................................................................... 91

Summary .................................................................................................................. 93

Appendix A: Storage Service Levels 95

Overview .................................................................................................................. 96

Storage service levels ............................................................................................... 96

Storage levels consumable by tenants ................................................................. 96

Storage service-level offerings ............................................................................. 96

Contents


Figures Figure 1. Key components of the EMC VSPEX solution for IaaS with VMware ...... 14

Figure 2. Logical architecture: VSPEX solution for IaaS with vCloud Suite ........... 20

Figure 3. EMC VNX sub-LUN tiering of production LUN data across two disk types23

Figure 4. Organization mappings for storage service levels ................................ 24

Figure 5. Storage DRS - Aligning datastore clusters with storage service levels .. 26

Figure 6. Architecture overview of vCOps vApp including EMC Storage Analytics 27

Figure 7. EMC VSI installed and available under Solutions and Applications ...... 36

Figure 8. EMC VSI Feature Manager .................................................................... 37

Figure 9. EMC VSI: VNX Array Discovery .............................................................. 38

Figure 10. EMC VSI : Block storage details and properties .................................... 38

Figure 11. EMC VSI : File storage details and properties ....................................... 39

Figure 12. Set multipathing policy ........................................................................ 40

Figure 13. Select NMP and PowerPath/VE multipathing policies for VNX devices . 40

Figure 14. Viewing VAAI support in the vSphere Client for VNX storage ................ 41

Figure 15. Start Profile Driven Storage service on vCenter server .......................... 42

Figure 16. VM Storage Profiles in vSphere Client .................................................. 43

Figure 17. Add VNX Provider details ..................................................................... 44

Figure 18. New EMC VNX vendor provider added – VNX5600 ............................... 44

Figure 19. Storage capabilities presented to vCenter by VNX Provider .................. 45

Figure 20. Add user-defined storage capability for VNX storage ........................... 46

Figure 21. Assign user-defined storage capability to datastore cluster ................. 46

Figure 22. System and user-defined storage capabilities displayed in vSphere .... 47

Figure 23. EMC Storage Analytics with VMware vCenter Operations Manager ....... 51

Figure 24. Launch EMC VSI storage provisioning wizard from vSphere Client ....... 54

Figure 25. Summary of vSphere tasks executed by EMC VSI when provisioning storage ................................................................................................ 55

Figure 26. 200 GB VMFS datastore ‘EPC_Prod2_Fin_LUN11’ available as a storage resource .............................................................................................. 55

Figure 27. Path management details for newly created 200 GB LUN/datastore .... 56

Figure 28. Datastore cluster fully configured in vSphere ....................................... 57

Figure 29. Endpoint configurations for vCAC ........................................................ 59

Figure 30. Enterprise groups created and managed by vCAC cloud administrator . 60

Figure 31. Enterprise groups resource defined vCAC cloud administrator ............. 60

Figure 32. Storage resource reservations as managed by the vCAC enterprise administrator....................................................................................... 61

Figure 33. Assigning cost profiles ........................................................................ 62

Figure 34. Provisioning groups managed by vCAC enterprise administrator.......... 62

Figure 35. Blueprint approval policy as set by provisioning group manager ......... 63

Contents


Figure 36. Create new VMware vSphere virtual machine blueprint ....................... 64

Figure 37. Create new blueprint ........................................................................... 65

Figure 38. Select vSphere template to clone from ................................................ 66

Figure 39. Complete build information for new blueprint ...................................... 66

Figure 40. Select custom build profiles for virtual machine .................................. 67

Figure 41. Enable virtual machine operations for cloud user ................................ 68

Figure 42. Deploy new virtual machine blueprint from self-service portal ............. 68

Figure 43. Reconfigure virtual machine resources from self-service portal ............ 69

Figure 44. Set execution options for virtual machine reconfiguration ................... 69

Figure 45. Decommissioning a virtual machine .................................................... 70

Figure 46. Executive Summary Virtual for vCAC virtual machines .......................... 71

Figure 47. Manage compute resource cost and storage cost profiles .................... 72

Figure 48. Manage compute resource cost and storage cost profiles .................... 72

Figure 49. Edit vCAC compute resource cost and storage cost profiles ................. 73

Figure 50. Chargeback by group allocated resources ........................................... 74

Figure 51. Chargeback by group by reservation .................................................... 75

Figure 52. Chargeback by owner by allocated resources ...................................... 76

Figure 53. Dashboard for reclaimed virtual machines in the enterprise private cloud ................................................................................................... 77

Figure 54. Reclamation Savings by Owner report .................................................. 77

Figure 55. vSphere UI dashboard high level overview .......................................... 78

Figure 56. Health dashboard showing immediate issues in environment ............. 80

Figure 57. End to end fault analysis covering actual fault, causes, and solution details ................................................................................................. 81

Figure 58. VMware vCOps root cause analysis solution details ............................ 81

Figure 59. EMC VNX Storage Topology dashboard ................................................ 83

Figure 60. ESA – EMC Storage Metrics displaying VNX LUN metrics ...................... 84

Figure 61. ESA – VNX Overview dashboard displayed object heat maps ............... 85

Figure 62. Virtual Machine Capacity details for management cluster .................... 86

Figure 63. Specify virtual machine configuration for What-If scenario ................... 86

Figure 64. What-If scenario – Adding ten new virtual machines ............................ 87

Figure 65. Edit policy to specify thresholds for underutilization of virtual machines87

Figure 66. List of oversized virtual machines with suggested optimal resource configurations ..................................................................................... 88

Figure 67. Schedule a report in vCOps or run the report manually ........................ 89

Figure 68. Successfully completed reports available for download ...................... 89

Figure 69. Storage service-level positioning for VSPEX management and orchestration ....................................................................................... 96

Figure 70. Storage service levels on VNX systems ................................................ 97

Figure 71. Storage service-level mapping from applications to storage array ........ 98

Contents


Tables Table 1. Terminology......................................................................................... 12

Table 2. Solution software ................................................................................ 17

Table 3. Deployment process stages and procedures ....................................... 34

Table 4. Management vCenter requirements ..................................................... 91

Table 5. vCAC Up to 1,000 machines (physical or virtual) .................................. 92

Table 6. vCOps hardware requirements............................................................. 92

Table 7. Production vCenter requirements ........................................................ 93

Table 8. Storage service levels example ............................................................ 97

Introduction


Chapter 1 Introduction

This chapter presents the following topics:

Introduction ............................................................................................................. 10

Purpose of this guide ............................................................................................... 11

Business value ......................................................................................................... 11

Scope ....................................................................................................................... 12

Audience .................................................................................................................. 12

Terminology ............................................................................................................. 12

Introduction


Introduction

EMC has joined forces with the industry-leading providers of IT infrastructure to create a complete virtualization solution that accelerates deployment of cloud infrastructure. Built with the best available technologies, VSPEX enables faster deployment, more simplicity, greater choice, higher efficiency, and lower risk. Validation by EMC ensures predictable performance and enables customers to select technology that uses their existing IT infrastructure while eliminating planning, sizing, and configuration burdens. VSPEX provides a proven infrastructure for customers who want to gain the simplicity that is characteristic of truly converged infrastructures and have more choice in individual solution components.

This EMC® VSPEX® cloud management solution is designed to enable customers to implement an on-site “Infrastructure as a Service” (IaaS) offering for their internal users. This solution is designed and implemented on a VSPEX proven infrastructure and hence uses the same hardware and software resources as defined in the VSPEX proven infrastructure collateral. Along with the software mentioned in the EMC VSPEX Private Cloud VMware vSphere 5.5 for up to 1,000 Virtual Machines Proven Infrastructure Guide, this solution also uses additional cloud management software to enable delivery of IT as a service. As more organizations attempt the transformation from the traditional “IT-as-a-cost-center” model to a more mature model of acting as a “service broker,” IaaS is the area where most customers focus their initial efforts.

This EMC VSPEX solution for IaaS deploys the best of EMC and VMware products and services to empower our customers to accelerate the implementation and adoption of private cloud while still enabling customer choice for the compute and networking infrastructure within the data center. In this solution, EMC is catering to those organizations that are looking for ways to further leverage their existing infrastructure investments and expand their use to cloud computing solutions.

EMC recommends reading related documents before proceeding.

VSPEX Proven Infrastructure Solution

Refer to this VSPEX Proven Infrastructure document: EMC VSPEX Private Cloud VMware vSphere 5.5 for up to 1,000 Virtual Machines Proven Infrastructure Guide.

vCloud Suite

Refer to the VMware vCloud Suite website for details about vCloud Suite.

Overview

Essential reading

Introduction


Purpose of this guide

The purpose of this document is to demonstrate how to implement an IaaS environment on an existing VSPEX Private Cloud solution using management and orchestration tools from VMware. This document provides an introduction to the VMware vCloud Suite and EMC VSPEX solutions, and specifically demonstrates integrations that exist in a VSPEX Private Cloud solution. Use this document as a reference to begin the planning and design process for your management and orchestration deployment, and to set the stage for successful implementation efforts.

Be aware that this particular design represents one of many possible ways to deploy an IaaS platform on a VSPEX Private Cloud environment. We chose this design because of the level of maturity of the toolsets and the ability to meet the use case requirements within a reasonable amount of time, maximizing the use of out-of-box capabilities.

Focusing on IaaS, this document defines the key elements needed to create an operational service offering, explains the processes required to implement the infrastructure and software stacks, and provides step-by-step directions to fulfill the primary requirements of a private cloud. This solution document focuses on the implementation of the vCloud Suite, and defines the core infrastructure requirements that must be met, such as storage, network, compute, and vSphere configurations, before you can layer on the private cloud components.

This document includes the most common use cases for implementing an IaaS-focused private cloud and provides direction and enablement content for foundational capabilities and requirements.

Business value

Business applications are moving into consolidated compute, network, and storage environments. EMC VSPEX solutions reduce the complexity of configuring every component of a traditional manual model. VSPEX solutions unify administration and self-service while maintaining security and process separation. This solution guide addresses the following business needs:

Automated provisioning through self-service portal

Automated monitoring

Multitenancy

Resource elasticity

Metering and chargeback

Performance monitoring

Storage resource management and analytics

Introduction


Scope

The guide assumes that the reader is familiar with VSPEX private cloud solutions and that a VSPEX private cloud architecture already exists in the customer environment. This guide describes the high-level steps required to deploy vCloud Suite on a VSPEX private cloud with VMware vSphere and EMC VNX, and it provides guidance for using vCloud Suite components to simplify and efficiently manage the VSPEX private cloud.

Audience

This document is targeted at partners, technical architects, and cloud solution engineers with a background in EMC and VMware solutions who are considering an IT-as-a-Service transformation and want to align the daily operational tasks of management and monitoring the private cloud with the skill sets required to ensure success. You should have an understanding of VMware vCloud and vSphere solutions, and be familiar with the EMC VNX storage platform solutions that you are planning to use as the enabling platforms hosting the IaaS offering.

You should also have a clear understanding of the provisioning and operational processes that are implemented within your own environments and recognize your end users’ general requirements for an IaaS solution. These requirements should include areas such as security, compliance, monitoring, management, data protection, and disaster recovery. Having these competencies will ensure a more relevant connection between the capabilities provided by this solution and the operational requirements of your organization that will ultimately bring additional business value.

Terminology

Table 1 lists terminology used in this guide.

Table 1. Terminology

Term Definition

FAST™ VP Fully Automated Storage Tiering for Virtual Pools

PaaS Platform as a Service

VAAI VMware vStorage APIs for Array Integration

VASA VMware vSphere Storage APIs for Storage Awareness

VSI Virtual Storage Integrator

VXLAN Virtual Extensible LAN

Solution Overview


Chapter 2 Solution Overview


Overview .................................................................................................................. 14

Key components ...................................................................................................... 14

Solution Overview


Overview

This EMC VSPEX solution for IaaS built on top of a VSPEX private cloud architecture enables customers to build a scalable multitenant platform for complete management and orchestration. This private cloud solution provides on-demand access and control of applications, servers, storage, and security while maximizing asset utilization. This solution uses vCloud Suite as the cloud management software and VSPEX private clouds as the physical infrastructure. This solution runs on the VMware vSphere virtualization layer backed by highly available EMC VNX® family storage components. The compute and network components, which are defined by the VSPEX partners, are designed to be redundant and are sufficiently powerful to handle the processing and data needs of the virtual machine environment.

This chapter introduces the components of vCloud Suite that enable management and orchestration in a VSPEX private cloud environment.

Key components

This section presents the key components used in this solution to enable complete management of a VSPEX stack through vCloud Suite. This section also explains integration points for EMC and vCloud Suite components that simplify the management and orchestration tasks.

EMC VNX

VMware vSphere

EMC VSI EMC PowerPath VE

VMware vCenter

VAAIVASA

vCenter Operations

VMware vCAC

EMC Storage Analytics

vCloud Networking and Security

Figure 1. Key components of the EMC VSPEX solution for IaaS with VMware

Solution Overview


VMware vCloud Suite is an integrated solution for building and managing a complete cloud infrastructure that meets IT’s most critical needs. vCloud Suite fulfills the promise of the software-defined datacenter by pooling industry-standard hardware and running each layer of the datacenter as a software-defined service. It creates pools of servers, storage, and networking with dynamically configurable security, availability, and management services that can meet the needs of all applications. The built-in self-service portal and catalog, policy-based infrastructure and application provisioning, and automated operations management help to complete the picture.

VMware vSphere

VMware vSphere transforms the physical resources of a computer by virtualizing the CPU, RAM, hard disk, and network controller. This transformation creates fully functional virtual machines that run isolated and encapsulated operating systems and applications like physical computers.

VMware vCenter VMware vCenter is a centralized management platform for the VMware Virtual Infrastructure. This platform provides administrators with a single interface for all aspects of monitoring, managing, and maintaining the virtual infrastructure, and can be accessed from multiple devices.

VMware vCenter also manages some advanced features of the VMware virtual infrastructure such as VMware vSphere High-Availability and DRS, along with vMotion and Update Manager.

vCenter Orchestrator VMware vCenter Orchestrator is an IT process automation engine that helps automate your cloud and integrate the VMware vCloud Suite with the rest of your management systems. Orchestrator saves time, removes manual errors, reduces operating expenses, and simplifies IT management. Orchestrator allows administrators and architects to develop complex automation tasks within the workflow designer, then quickly access and launch workflows directly from within the vSphere Client or by using various triggering mechanisms.

vCloud Automation Center

vCloud Automation Center enables customized, self-service provisioning and lifecycle management of cloud services that comply with established business policies. It provides a secure portal where authorized administrators, developers, or business users can request new IT services and manage existing computer resources from predefined user-specific menus.

vCenter Operations Manager

VMware vCenter Operations Manager is the key component of the vCenter Operations Management Suite. It provides a new, simplified approach to the operations management of vSphere’s physical and cloud infrastructure. Using patented, self-learning analytics and an open, extensible platform, vCenter Operations Manager provides you with operations dashboards to gain deep insights and visibility into the health, risk, and efficiency of your infrastructure, providing performance management and capacity optimization capabilities.

VMware vCloud Suite

http://www.vmware.com/solutions/datacenter/software-defined-datacenter/index.html

http://www.vmware.com/products/datacenter-virtualization/vcenter-operations-management/overview.html

http://www.vmware.com/products/datacenter-virtualization/vcenter-operations-management/overview.html

Solution Overview



VMware vCloud Networking and Security (vCNS) is the leading software-defined networking and security solution. It enhances operational efficiency, unlocks agility, and provides the extensibility to rapidly respond to business needs. It provides a broad range of services in a single solution, including virtual firewall, VPN, load balancing and Virtual Extensible LAN (VXLAN) extended networks.

The EMC VNX storage system is a powerful, trusted, and smart storage array platform that provides a high level of performance, availability, and intelligence in organizations within the business.

VSPEX private cloud customers can leverage the advanced storage tiering features and efficiencies of the VNX storage array to deliver multiple storage service levels to their various organizations, accelerating and simplifying their as-a-service offerings in their private cloud environment.

Virtual Provisioning

EMC Virtual Provisioning™ can reduce cost, improve capacity utilization, and simplify storage management. Users can present a large amount of capacity to a host and then consume space only as needed from a shared pool, reducing initial over-allocation of storage capacity. Virtual Provisioning can reduce labor costs by simplifying data layout and reducing the steps required to accommodate capacity growth.

Fully Automated Storage Tiering

EMC Fully Automated Storage Provisioning for Virtual Pools (FAST™ VP) for VNX optimizes efficiency across all drive types in the array to improve system performance while reducing cost. FAST VP dynamically allocates workloads based on the configured service level and nondisruptively moves workloads across drive types, optimizing overall system performance. FAST VP moves the most active parts of the workload to high-performance flash disks and the least frequently accessed data to lower-cost drives, leveraging the best performance and cost characteristics of the different drive types.

Unisphere

EMC Unisphere® is an intuitive management interface that allows IT managers to dramatically reduce the time required to provision, manage, and monitor storage assets. Unisphere delivers the simplification, flexibility, and automation that accelerate the transformation to the private cloud.

EMC provides tight integration points for vCloud suite which simplify management and provisioning of storage. vCloud Suite also provides the capability to monitor the health and performance of storage resources on the VNX array from within.

EMC Virtual Storage Integrator

EMC Virtual Storage Integrator (VSI) is the free vCenter plug-in provided by EMC that extends the vCenter UI to add additional EMC-specific capabilities. EMC VSI provides multiple feature sets including Storage Viewer (SV), Path Management, and Unified

EMC VNX

EMC Integration for vCloud Suite

Solution Overview


Storage Management. Unified Storage Management simplifies the provisioning of VNX virtual pooled storage for the private cloud.

VMware vStorage APIs for Array Integration (VAAI)

EMC VNX supports VMware vStorage APIs for Array Integration (VAAI), which offloads virtual machine storage operations to the array to optimize server performance. In the VMware environments, the following VAAI components are supported:

VMware vSphere Storage APIs for Storage Awareness (VASA)

VASA allows VMware administrators to view and classify, from a single location VNX drive types in a more sophisticated way, using the storage vendors VASA provider to interrogate the storage and classify it in terms of its attributes and underlying technology. This can range from basic disk type to the RAID level, automated tiering capabilities, and replication status.

EMC Storage Analytics

EMC Storage Analytics can be used as a plug-in with VMware’s vCenter Operations Manager to provide a powerful management tool for VMware and storage administrators to access real-time intelligent analytics for their VNX platform. Obtain detailed statistics via customizable dashboards, heat maps, and alerts while also accessing topology mapping in a VMware environment.

This solution is built on top of a VSPEX configuration and assumes that the required hardware is installed and set up within the business according to the configuration guidelines in this VSPEX Proven Infrastructure document: EMC VSPEX Private Cloud VMware vSphere 5.5 for up to 1,000 Virtual Machines Proven Infrastructure Guide.

Table 2 lists the software requirements for this solution.

Table 2. Solution software

Software Configuration Notes

VMware vCloud Automation Center (vCAC)

5.2.3 VMware vCloud Automation Center

VMware vSphere ESXi 5.5 Build 1106514 Server hypervisor

VMware vCenter Server 5.5 Build 1312299 vSphere Management Server

VMware vCenter Operations Manager (vCOps)

5.7.1 VMware vCenter Operations Management


5.1 Networking and security for vCloud site

Microsoft SQL Server o 2012 Enterprise Edition Database server for vCenter, vCAC

Microsoft Windows Server 2008 R2 SP1, 2012 R2 Operating system for the server environment

EMC VSI 5.6 EMC storage plug-in for VMware vSphere Client

EMC Storage Analytics 2.0 EMC Storage Analytics Adapter

EMC Unisphere 1.6.0.8 Management software for EMC VNX

Hardware list

Software list

Solution Overview


Software Configuration Notes

EMC VNX OE for file o 8.1 Operating environment for file

EMC VNX OE for block o OE Release 33 Operating environment for block

EMC PowerPath Virtual Edition

5.8 Multipathing and load balancing for block access.

Solution Design


Chapter 3 Solution Design


Overview .................................................................................................................. 20

Solution architecture ............................................................................................... 20

Storage design ......................................................................................................... 21

Solution Design


Overview

It is important to understand the roles of the various solution components that are used to enable management and orchestration of the VSPEX Private Cloud. This section explains the design and layout of this cloud management solution that was implemented on a VSPEX private cloud architecture to deliver infrastructure as a service. The functions served by the different components along with their use cases in this solution are discussed in this section.

Solution architecture

This section describes the environment and supporting infrastructure for this EMC VSPEX solution for IaaS.

Figure 2. Logical architecture: VSPEX solution for IaaS with vCloud Suite

High-level solution architecture

Solution Design


The management infrastructure for this private cloud solution is critical to the availability of its supporting components. This solution has two management layers which separate the virtual infrastructure into two tiers of resource management:

Management cluster

Production resource clusters

Management cluster

The management cluster hosts all virtual machines used for managing the cloud infrastructure. This cluster supports the components responsible for functions such as the user portal and automated provisioning, monitoring, networking, security, and metering.

All server and virtual machine components within the management cluster are managed by a separate, higher level vCenter Server instance. This instance can be located in the management cluster, but for highest possible availability, should be located on a separate set of infrastructure servers.

The management cluster is supported by three VMware vSphere ESXi servers, configured in a vSphere cluster using VMware vSphere Distributed Resource Scheduler (DRS) and VMware vSphere High Availability (HA). This vSphere cluster uses server, network and storage resources of its own, separate from the production resource clusters. Fibre Channel connectivity between the management cluster and the VNX array is not a mandatory requirement; iSCSI or NFS connectivity can also be used to provide storage to the management cluster if preferred. All storage is RAID protected and all ESXi servers run EMC PowerPath®/VE for automatic path management and load balancing.

Production resource clusters

The production resource clusters host all systems used by different organizations to support the business needs of the enterprise customer. These resource clusters are managed by a second vCenter Server instance located in the management cluster, which serves as the vSphere endpoint for vCloud Automation Center (vCAC).

Add server, network, and storage resources to the existing cluster, and change resource reservations at the vCAC layer to make the additional resources available for consumption. Extend existing storage resources automatically using EMC VSI. All storage configured to support the production resource clusters is RAID protected and all ESXi servers run EMC PowerPath®/VE for automatic path management and load balancing.

Sizing considerations for the production resource clusters will vary across enterprise customers. Base the sizing considerations on the expected workload for each organization.

Storage design

The VNX storage array provides many features that allow provisioning of IaaS. Since this solution is built on top of a VSPEX Private Cloud, it uses the same storage configuration as mentioned in the VSPEX Private Cloud proven infrastructure

Solution Design


document. As a result, this solution uses a single storage service level as the base storage configuration to support up to 1000 reference virtual machines.

For more information on reference virtual machines refer to:

EMC VSPEX Private Cloud VMware vSphere 5.5 for up to 1,000 Virtual Machines Proven Infrastructure Guide

Additional storage service levels can be created within a VNX array to achieve different levels of performance and most effectively align resources with performance requirements. Refer to Appendix A for more information on creating additional storage service levels.

Note: The two additional storage service level examples in Appendix A fall outside of the guidelines in the underlying Private Cloud Proven Infrastructure. If a specific number of reference virtual machines will be supported, use only the storage configuration mentioned in the VSPEX Proven Infrastructure document. If virtual machines are deployed on other storage service level examples provided in Appendix A then the number of reference virtual machines supported may be lower.

As validated in the VSPEX Proven Infrastructure, this solution uses EMC’s FAST VP technology which enables storage tiering within the storage service level. Storage tiering is the assignment of data to different types (tiers) of storage media to reduce total storage cost. FAST VP makes automatic storage tiering possible on VNX storage arrays. FAST VP operates by periodically relocating the most active data to the highest performance storage tier, while relocating less active data to the lower performing storage tiers, as appropriate, when new data is promoted.

FAST VP uses intelligent algorithms to continuously analyze devices at the sub-LUN level. This enables it to identify and relocate the specific parts of a LUN that are most active and would benefit from being moved to higher-performing storage such as enterprise flash drives. It also identifies the least active parts of a LUN and relocates that data to higher-capacity, more cost-effective storage. Performance measurement and user-defined policies determine data movement between tiers. FAST VP moves the data automatically and non-disruptively.

The VSPEX Private Cloud uses two disk types within the storage pool with each providing a different tier of performance and capacity, as shown in Figure 3.

VNX storage tiering and FAST VP

Solution Design


Figure 3. EMC VNX sub-LUN tiering of production LUN data across two disk types

Figure 4 shows how the storage pool and LUNs are mapped to different enterprises within the business.

Production clouds are organizational clouds for virtual machines with maximum performance requirements.

Test/Development clouds are clouds are organization clouds for virtual machines with medium performance requirements.

Archive/Low Cost clouds are organizational clouds for virtual machines with least performance requirements.

Storage mapping

Solution Design


Finance-ESXi

Mfg-ESXi

Prod-2

LUNs

Production Clouds

VM01 VM03VM02

Test/Development Clouds

VM01 VM03VM02

Archive/Low Cost Cloud

VM01 VM03VM02

Storage Service Level

Organization Mappings for Storage Service Levels EMC VNX

FlashDrives

SAS Drives

Figure 4. Organization mappings for storage service levels

vCloud Suite design

The combination of standardization, workflow automation, and self-service gives businesses the agility they need to offer IT as a service on virtual infrastructures at reduced cost. The Management and Orchestration Workflow Automation solution combines the converged infrastructure of VSPEX with service catalog and process automation capabilities of vCloud Suite that enable IT organizations to deliver services rapidly, efficiently, and cost-effectively.

vCloud Suite components possess attributes which collectively are necessary for successful management and orchestration of a Private Cloud. These components along with EMC VNX and EMC integration software for vCloud Suite enable provisioning of Infrastructure as a service on a VSPEX Private Cloud.

VMware vSphere is the leading server virtualization platform in the industry. VMware vSphere provides flexibility and cost savings to end users by consolidating large inefficient server farms into nimble reliable cloud infrastructures. The core components of VMware vSphere are the vSphere Hypervisor and the vCenter Server.

Use case

vCenter Server This solution uses VMware vCenter for system management. Physical servers in this solution are clustered to provide high availability. This solution uses two vCenter Servers. The first vCenter server manages the management cluster. The second vCenter Server manages the production cluster which is used as an endpoint for consumable resources in vCloud automation center.

VMware vSphere

Solution Design


This solution takes advantage of VMware Storage API for Array Integration to offload virtual machine storage operations to the storage array. It uses VMware Storage API for storage awareness which provides a single management interface to view VNX drive types.

vCenter uses the EMC VSI plug-in that provides features like Storage Viewer, Path Management and Unified Storage Management. Unified Storage Management simplifies provisioning of VNX pooled storage through vCenter for various organizations within the business.

EMC VSI Because VSI is modular in design, administrators can add, remove and update features independently providing a flexible customized user experience. Features available for VSI include Storage Viewer, Path Management, Symmetrix storage replication adaptor (SRA), Unified Storage Management, AppSync Management, and EMC RecoverPoint Management.

The three features used in this solution are:

Storage Viewer feature extends the VMware vSphere Client to facilitate the discovery and identification of VNX storage devices that have been provisioned to ESXi hosts and virtual machines. Storage Viewer presents the underlying storage details to the vSphere administrator by merging the data of several different storage mapping tools into a few seamless VMware vSphere Client views.

Path Management feature for VMware Native Multipathing and EMC PowerPath/VE provides a mechanism for changing the multipath policy for groups of LUNs based on storage class and virtualization objects.

Unified Storage Management feature provides array-based storage management and provisioning for VMAX, VNX, VNXe, CLARiiON, Celerra arrays. This feature also allows for virtual machine decompression, compression, cloning and extending datastore capacity.

VMware vSphere Storage DRS To prevent hot spots and over- or under-utilized datastores, VMware vSphere Storage DRS provides smart virtual machine placement and load-balancing mechanisms based on I/O and space capacity. Datastore clusters form the basis of Storage DRS. As with a cluster of ESXi hosts, a datastore cluster is used to aggregate storage resources, enabling smart and rapid placement of the virtual disk files of a virtual machine and load balancing of existing workloads.

This solution for the VSPEX Private Cloud aligns storage with datastore clusters. All storage devices within a datastore cluster must have consistent size and performance characteristics. Storage assigned to an organization in VMware vCAC consists of devices which have been aggregated to a single unit of consumption—a single datastore cluster—in vSphere. Though vCAC v5.2 is unaware of Storage DRS (SDRS) operations, all SDRS movements of virtual machine files are contained within a single datastore cluster, and all LUNS within each datastore cluster in this solution are available to only one tenant; therefore the scope of virtual machine file movement is contained within the tenant’s resources.

Solution Design


In the example shown in Figure 5 for the Manufacturing organization, all of the storage devices are aggregated into a single datastore cluster. Depending on the choice of service level when a blueprint in vCAC is provisioned, each cluster presents a single target for the provisioning of virtual machines in vCAC blueprints.

Figure 5. Storage DRS - Aligning datastore clusters with storage service levels

VMware vCloud Automation Center (vCAC) enables you to rapidly deploy and provision business-relevant cloud services across your private cloud and physical infrastructure. Acting as a service governor, it provides a cross-cloud storefront for IaaS and platform as a service (PaaS) deployments. It empowers organizations to enforce business and IT policies throughout the service life cycle, helping them to transform virtualized environments into software-defined cloud datacenters.

Use case

vCAC provides the ability to carve a shared infrastructure up into logical units and logical capacities that can be handed over to different business units. It does this with machine blueprints. Cloud users can choose from a self-service catalog of custom defined blueprints, each containing specific resources appropriate to different applications or business units.

vCAC is designed to leverage physical components such as those that comprise the VSPEX proven Infrastructure. It supports the differing requirements of an enterprise’s many business units and integrates with a wide variety of existing IT systems and best practices. Through the workflow designer, vCAC can easily invoke vCenter Orchestrator workflows to extend lifecycle state transitions and machine command menus, resulting in entire processes being automated without requiring any manual cycles from IT teams.

A number of user roles and responsibilities are defined and used in the structure of vCAC. The administration of users and compute resources in vCAC is managed through the vCAC console, which is the administrative portal. The relevant groups are summarized below:

Cloud administrator The cloud administrator configures resource endpoints and enterprise groups, where enterprise group administrators and their respective compute resources are defined.

vCloud Automation Center

Solution Design


Enterprise groups Enterprise groups contain enterprise administrators that manage their respective cloud resources, as defined by the cloud administrator. Enterprise administrators are responsible for configuring and creating provisioning groups for individual departments as well as resource reservations and customizations for network, storage, compute and cost profiles. Approval groups and policies are also defined by the enterprise administrator.

Provisioning groups The users in these provisioning groups are the users and consumers of the infrastructure provided to them by their enterprise administrator. The enterprise administrator assigns compute, storage and network resources and makes them available to the provisioning group manager.

VMware vCenter Operations Manager provides an integrated approach to performance, capacity, and configuration management. This solution uses analytics to provide the intelligence and visibility required to proactively ensure service levels in virtual and cloud environments.

vCenter Operations Manager provides pre-built and configurable dashboards for real-time performance, capacity and configuration management.

Figure 6. Architecture overview of vCOps vApp including EMC Storage Analytics

vCenter Operations Manager

Solution Design


As shown in Figure 6, vCenter Operations Manager is distributed as a vApp which you can import and deploy to vSphere. This vCenter Operations Manager vApp is located in the Management Cluster of this Enterprise private cloud solution and consists of two virtual machines:

UI virtual machine

Analytics virtual machine

UI virtual machine The UI virtual machine allows you to access the results of the analytics in the form of badges and scores using the web-based application for the vSphere UI virtual machines.

vCenter Operations Manager Administration Portal The vCenter Operations Manager Administration Portal provides a user interface for vCenter Operations Manager vApp maintenance and management tasks.

Analytics virtual machine The Analytics virtual machine is responsible for collecting data from vCenter Server, vCenter Configuration Manager, and third party data sources, such as metrics, topology, and change events

Capacity Collector The Capacity Collector collects metrics and computes derived metrics.

Use case

Performance data is abstracted to health, risk, and efficiency measures that allow IT to identify potential performance problems with less effort. Capacity analytics identify over-provisioned resources so they can be right-sized for more efficient use of virtualized resources. "What If" scenarios eliminate the need for spreadsheets, scripts, and rules of thumb.

Integrating vCenter Operations with the EMC Storage Analytics Suite enables full end-to-end visibility of the entire infrastructure from virtual machine to LUN and every point in between. This enables IT administrators to quickly visualize the health of the EMC arrays (both block and file) using a simple Performance-at-a-glance tab.

EMC Storage Analytics (ESA) links VMware vCenter Operations for storage with the EMC Adapters for VNX®. The vCenter Operations Manager displays performance and capacity metrics from EMC storage systems with data that the adapter provides by:

Connecting to and collecting data from block and file systems.

Converting the data into a format that vCenter Operations can process.

Passing the data to the vCenter Operations collector.

vCloud Networking and Security enables the creation of VXLANs which enable creation of layer 2 logical networks without the help of VLANs. VXLANs, network services like NAT, and a DHCP-application-level firewall along with data security, help in segregating tenant traffic and providing multitenancy.


Solution Design


Use case

Through vCAC, vCloud networking and security is specified as a vSphere endpoint. This enables vCAC to discover network resources like load balancers, network paths, and security groups. These network paths can be added to reservations while load balancers and security groups can be added to blueprints.

Solution Design


Design considerations and best practices

Follow the design considerations and best practices in this section to efficiently deliver IaaS to different organizations and to simplify management for system and cloud administrators.

Because this solution is built on top of the VSPEX private cloud solution for vSphere, the core network design considerations are the same as those detailed in the following Proven Infrastructure document:

EMC VSPEX Private Cloud VMware vSphere 5.5 for up to 1,000 Virtual Machines Proven Infrastructure Guide

Because a self-service-based private cloud environment has separate tiers and levels of users, certain network practices should be deployed to simplify management. This solution uses VMware vSphere Distributed Switch (VDS) to simplify virtual machine network configuration. For more information on VDS refer to the VMware website.

VXLAN-based networking through vCNS

Create a VMware vSphere distributed switch (VDS) within the datacenter datacenter for clusters belonging to each enterprise. Add the uplink ports meant for virtual machine traffic to the VDS.

Configure the clusters and VDS for VXLAN networking.

Create virtual wires for each Production, Test and Development, and Archive production group.

VLAN-based networking

Create a VMware vSphere distributed switch (VDS) within the datacenter for clusters belonging to each enterprise. Add the uplink ports meant for virtual machine traffic to the VDS

Create a separate Distributed Virtual (DV) port group for each Production, Test and Development, and Archive provisioning group deployed within the business. Each DV port group should have a dedicated VLAN.

Note: If multiple NICs are available, use the fastest network adapters for Production provisioning groups and the slower adapters for Archive provisioning groups. Configure the settings for a DV port group so that it uses only specific network adapters for traffic.

Because this solution is built on top of the VSPEX private cloud solution for vSphere the storage design for the production-level storage is the same as detailed in that solution. Appendix A of this document details the design for two additional storage levels—the Test and Development storage level and the Archive storage level. The two additional storage levels mentioned in Appendix A are only provided as examples. VSPEX recommends using only the storage layout described in the VSPEX Private Cloud solution for vSphere. The use of additional storage tiers may impact the potential number of reference virtual machines supported by the Private Cloud Proven Infrastructure.

Network design

Storage design

Solution Design


Storage layout examples

The Production, TestDev, and Archive provisioning groups of each enterprise use the storage configuration as outlined in the EMC VSPEX Private Cloud VMware vSphere 5.5 for up to 1,000 Virtual Machines Proven Infrastructure Guide.

Since the three clouds will be supporting different performance profiles, EMC FAST VP technology would automatically tier the data within the storage pool to achieve optimal performance.

To simplify storage management and to achieve separation of resources for each provisioning group, EMC recommends dedicating a particular LUN (or group of LUNs in a storage cluster) to a particular provisioning group only and not having different provisioning groups using the same LUN (or group of LUNs) for storage.

Virtualization design

For virtualization design considerations, refer to EMC VSPEX Private Cloud VMware vSphere 5.5 for up to 1,000 Virtual Machines Proven Infrastructure Guide.

To simplify management and efficiently utilize resources, follow these practices:

Create separate host clusters for each enterprise group (manufacturing, finance, accounting, and so on).

Create separate reservations for each provisioning group within the enterprise.

Note: You can create additional host groups under the parent host group if separation of Production, Test and Development, and Archive clouds is desirable.

Dedicate a particular LUN to a specific provisioning group only.

Solution Design


Solution Implementation


Chapter 4 Solution Implementation


Overview .................................................................................................................. 34

Installing and configuring vCloud Management cluster ........................................... 34

Deploying vCloud Networking and Security.............................................................. 47

Deploying vCAC ........................................................................................................ 48

Deploying vCenter Operations Manager ................................................................... 50



Overview

This chapter introduces the steps and processes involved in deploying the solution. Table 3 lists the main stages in the solution deployment process along with links to the relevant information and procedures.

Table 3. Deployment process stages and procedures

Stage Description

1 Install and configure vCloud Management Cluster

2 Deploy vCAC

3 Deploy vCloud Networking and Security

4 Deploy vCenter Operations Manager

5 Deploy vCenter Orchestrator

Installing and configuring vCloud Management cluster

Complete these tasks to install and configure a vCloud Management server:

1. Install the ESXi 5.5 hypervisor on the physical servers being deployed for the solution. Refer to the vSphere Installation and Setup Guide for more detail.

2. Configure ESXi networking including NIC trunking, VMkernel ports, and virtual machine port groups and Jumbo Frames. Refer to the vSphere Networking document for more detail.

3. Install and configure PowerPath/VE to manage multipathing for VNX LUNs (block only). Refer to PowerPath/VE for VMware vSphere Installation and Administration Guide for more detail (block storage only).

4. Connect the VMware datastores to the ESXi hosts deployed for the solution. Refer to the vSphere Storage Guide for more detail.

Complete these tasks to set up and configure a Microsoft SQL Server database for the solution:

1. Create two virtual machines to host SQL Server for the Management and Production vCenter instances separately. Verify the virtual server meets the hardware and software requirements based on the resource sizing guide in Chapter 6. Refer to the Microsoft Developer Network webpage for more detail.

Note: EMC recommends using SQL Server clusters for high availability.

2. Install Microsoft Windows Server 2008 R2 on the virtual machine created to host SQL Server. Refer to the Microsoft TechNet webpage for more detail.

3. Install SQL Server on the virtual machine designated for that purpose. Refer to the Microsoft TechNet webpage for more detail.

Setting up and configuring SQL Server database

http://pubs.vmware.com/vsphere-50/topic/com.vmware.ICbase/PDF/vsphere-esxi-vcenter-server-50-networking-guide.pdf

http://msdn.microsoft.com/

http://technet.microsoft.com/

http://technet.microsoft.com/



4. Create the database required for the vCenter server on the appropriate datastore. Refer to Preparing vCenter Server Databases for more detail.

5. Create the database required for Update Manager on the appropriate datastore. Refer to Preparing the Update Manager Database for more detail.

Complete these tasks to deploy management and production resource cluster vCenter Servers:

1. Create two virtual machines to be used for management and production resource cluster vCenter Server. Refer to the vSphere Virtual Machine Administration Guide for details.

2. Install Windows Server 2008 R2 Standard Edition on the two vCenter host virtual machines.

3. Install VMware Tools, enable hardware acceleration, and allow remote console access on the two vCenter host virtual machines. Refer to the vSphere Virtual Machine Administration Guide for details.

4. Create the 64-bit vCenter and 32-bit vCenter Update Manager ODBC connections on the two vCenter host virtual machines. Refer to vSphere Installation and Setup and Installing and Administering VMware vSphere Update Manager for details.

5. Install vCenter Server software on the two vCenter host virtual machines. Refer to vSphere Installation and Setup for details.

Complete these tasks to prepare and configure a vCenter management cluster:

1. Create a virtual datacenter for the management cluster. For details, refer to vCenter Server and Host Management.

2. Type the vSphere license keys in the vCenter licensing menu. For details, refer to vSphere Installation and Setup.

3. Connect management vCenter to management cluster ESXi hosts. For details, refer to vCenter Server and Host Management.

4. Create a management cluster and move the ESXi hosts into it. For details, refer to vSphere Resource Management.

5. Perform ESXi host discovery from the Unisphere console. For details, refer to Using EMC VNX Storage with VMware vSphere–TechBook.

This section provides guidance on integrating EMC VNX with VMware vSphere through VSI plug-in, VAAI and VASA.

EMC VSI for VMware

Refer to the VSI plug-in documentation for installing VSI.

EMC VSI for VMware vSphere: Unified Storage Management Product Guide

EMC VSI for VMware vSphere: Storage Viewer Product Guide

EMC VSI for VMware vSphere: Path Management

Deploying management and production resource cluster vCenter servers

Preparing a management cluster

EMC integration



Once VSI has been installed on the client machine, find the EMC VSI icon in vSphere located under vSphere Client -> Solutions and Applications, as shown in Figure 7.

Figure 7. EMC VSI installed and available under Solutions and Applications

Verify that all newly installed VSI features were installed and enabled by checking feature manager. In our solution we installed Path Manager, Storage Viewer, and Unified Storage Management as shown in Figure 8.



Figure 8. EMC VSI Feature Manager

As part of the VSI installation, Solutions Enabler 7.6 is installed. Also, for VSI to obtain and present PowerPath/VE multipath details for the underlying storage, install the PowerPath Remote Tools for Windows package. VSI currently supports PowerPath Remote Tools version 5.7.b173, or later.

Configure Solutions Enabler Configure Solutions Enabler by adding the hostname or IP address of the Solutions Enabler API server.

Note: If you do not specify remote API server it will default to ‘localhost’.

Discover VNX array To discover the VNX arrays, add the IP address of each storage processor and the login credentials and click discover new VNX system as shown in Figure 9.



Figure 9. EMC VSI: VNX Array Discovery

Once the storage arrays have been discovered, the vSphere admin can get a more granular view of existing datastores. Select a host and select the EMC VSI tab, then highlight the desired datastore as shown in Figure 10 and Figure 11.

Figure 10. EMC VSI : Block storage details and properties



Figure 11. EMC VSI : File storage details and properties

Unified Access Control Utility Overview EMC Unified Access Control provides a means to restrict vSphere administrators from provisioning storage on specific storage pools within a given array. By providing this control, storage administrators can build designated storage pools dedicated for vSphere administrators to provision storage as needed. Alternatively, the storage administrator can allow provisioning to all current and future pools.

VSI Path Management Before using Path Management, be sure that RTOOLS is installed on the host running VSI and that the host has appropriate licenses.

To manage paths through VSI:

1. Right-click the desired vSphere cluster, and select EMC > Set Multipathing Policy, as shown in Figure 12.



Figure 12. Set multipathing policy

2. Select the device type and desired policy from Select Multipathing Policy as shown in Figure 13.

Figure 13. Select NMP and PowerPath/VE multipathing policies for VNX devices



VMware vStorage APIs for Array Integration (VAAI)

To verify that VAAI is working correctly, from the Configuration tab, under Hardware -> Storage, find the column in the datastores view labeled Hardware Acceleration as shown in Figure 14. This column indicates the support status for the device or datastore. There are three possible values that can populate this column: Supported, Not Supported, or Unknown.

More detailed information on VMware VAAI with EMC VNX can be found in the VMware vStorage APIs for Array Integration with EMC VNX series for SAN White Paper.

Figure 14. Viewing VAAI support in the vSphere Client for VNX storage

VMware vSphere Storage APIs for Storage Awareness (VASA)

VASA is commonly used to enable profile driven storage where a virtual machine is attached to a storage capability and is then matched and placed on a suitable datastore of the same capability profile. This solution uses VASA functionality to organize datastores of identical attributes and capabilities and place them in datastore clusters, configured for each business organization, where they can take advantage of Storage DRS to maximize capacity utilization.

Configuring VASA Storage Provider on VNX The VASA provider for VNX is available by default, with no additional user configuration required.

1. Set up storage profiles on the vCenter server by starting VMware vSphere Profile Driven Storage Service in Windows Services as shown in Figure 15.

http://www.emc.com/collateral/hardware/white-papers/h8293-vaai-vnx-wp.pdf



Figure 15. Start Profile Driven Storage service on vCenter server

2. Once the agent is running, move the VM Storage profiles to the vSphere Client by selecting Home -> Management -> VM Storage Providers, as shown in Figure 16.



Figure 16. VM Storage Profiles in vSphere Client

3. Click Add, as highlighted in Figure 17, and enter the relevant details for the VNX storage array, including:

Array name and administrative login details

URL of the provider service on VNX, for example, https://<ipAddress>/vasa/Services/vasaService



Figure 17. Add VNX Provider details

4. Once the details are accepted and the array is successfully added, the array is displayed in the vendor provider list, as shown in Figure 18.

Figure 18. New EMC VNX vendor provider added – VNX5600



5. Specific VNX provider details and status are displayed under Vendor Provider Details, which can also be manually synchronized from this section.

6. View the complete list of storage capabilities which the VASA provider presents to the Storage Profiles service in vCenter by selecting Home-> Management->VM Storage Profiles and click Manage Storage Capabilities, as shown in Figure 19. This list is initially shows all system-defined storage capabilities which, depending on the storage type being presented, are automatically detected by VASA and displayed in vSphere as part of datastore details.

Figure 19. Storage capabilities presented to vCenter by VNX Provider

You can manually enter customized storage capabilities to more accurately reflect or represent a service or configuration that is not system-defined, as highlighted in the example in Figure 20. To add a user-defined storage capability, click Add within Manage Storage Capabilities, as shown in Figure 20.



Figure 20. Add user-defined storage capability for VNX storage

This user-defined storage capability can be assigned to single or multiple datastores. For this solution, it is assigned to all members of a datastore cluster, as shown in Figure 21.

Figure 21. Assign user-defined storage capability to datastore cluster

In the vSphere Client, you can view the system-defined and user-defined storage capabilities for each member of this datastore cluster, as shown in Figure 22, where the user-defined storage capabilities have been fully expanded into view.



Figure 22. System and user-defined storage capabilities displayed in vSphere

Deploying vCloud Networking and Security

In the private cloud, the VMware vCloud Networking and Security (vCNS) feature can provide firewall, VPN (Virtual Private network), Load Balancer, VXLAN, NAT (Network Address Translation), HA (High Availability), Data Security, Endpoint, and vCloud Ecosystem Framework.

Complete these tasks to deploy vCloud Networking and Security. More detail for performing each task is provided in the VMware vCloud Networking and Security Documentation.

Note: vCNS documentation and UI still use the vShield product name.

1. Install and configure vShield Manager. Deploy the OVF template downloaded from the VMware website and configure.

2. Configure the network settings of the vShield Manager. Assign the IP address, gateway, and DNS to the vShield Manager through the CLI (command line interface).

3. Log in to the vShield Manager User Console. Open a web browser and navigate to the IP address just assigned to the vShield Manager.

4. Set up vShield Manager. Specify settings for vCenter Server, DNS and NTP server, and Lookup server details.

5. Install the vShield license through vSphere Client.

https://www.vmware.com/support/pubs/vshield_pubs.html




Deploying vCAC

Complete the tasks described in this section to deploy the vCAC components.

VMware recommends that you deploy a dedicated server running Microsoft SQL Server to host the vCAC database. Create the database server by following these steps:

1. Prepare one vCAC authorization store. Choose one type of vCAC authorization store (SQL-based, Active Directory-based, or file-based) and create it. For more information, refer to the vCAC 5.2 Installation Guide.

2. Install Microsoft .Net Framework 4.5 in the database server.

3. Install and run the vCAC Prerequisite Checker for Database components.

4. Create the vCAC database by using the vCAC wizard.

Install the vCAC web server by following these steps:

1. Create a virtual machine to be used for the vCAC web server. For more information, refer to the vSphere Virtual Machine Administration Guide.

2. Install the vCAC web server guest operating system, on the vCAC web server virtual machine. Refer to the vCloud Automation Center Support Matrix.

3. Update the virtual machine by installing VMware Tools, enabling hardware acceleration, and allowing remote console access. For more information, refer to the vSphere Virtual Machine Administration Guide.

4. Install Microsoft .Net Framework 4.5.

5. Install the IIS Server Role with role services as specified in Installing IIS 7 on Windows Server 2008 or Windows Server 2008 R2,

6. Install and run the vCAC Prerequisite Checker for web components. For more information, refer to the vCAC 5.2 Installation Guide.

7. Install vCAC web components. Install and configure the administration website, reports website, and the model manager.

Set up the vCAC server Manager Service by following these steps:

1. Create a virtual machine to be used for the vCAC server Manager Service.

2. Install Windows Server 2008 R2 SP1, which is the vCAC server guest operating system, on the vCAC server virtual machine.

3. Update the virtual machine by installing VMware Tools, enabling hardware acceleration, and allowing remote console access.


5. Install the IIS Server Role with role services as specified in Installing IIS 7 on Windows Server 2008 or Windows Server 2008 R2.

6. Install and run the vCAC Prerequisite Checker for Manager Service and DEM Worker and Orchestrator Service components.

Database server

vCAC web server

vCAC server Manager Service

http://www.vmware.com/pdf/vcac-52-installation-guide.pdf

http://www.vmware.com/pdf/vcac-52-support-matrix.pdf

http://www.iis.net/learn/install/installing-iis-7/installing-iis-7-and-above-on-windows-server-2008-or-windows-server-2008-r2







7. Install and configure the vCAC Manager Service.

8. Install the DEM Orchestrator in the same virtual machine.

Create a DEM worker server by following these steps:

1. Create a virtual machine to be used for the DEM Worker server.

2. Install the guest operating system on the DEM Worker server virtual machine. Refer to the vCloud Automation Center Support Matrix

3. Install VMware Tools, enable hardware acceleration, and allow remote console access.


5. Install and run the vCAC Prerequisite Checker for DEM Worker and Orchestrator Service components.

6. Install the DEM Worker in the virtual machine.

7. Create a virtual machine to be used for the vCAC Proxy Agent Machine.

8. Install the guest operating system on the vCAC Proxy Agent Machine virtual machine. Refer to the vCloud Automation Center Support Matrix.



11. Install and run the vCAC Prerequisite Checker for DEM Worker and Orchestrator Service components.

12. Install the DEM Worker in the virtual machine.

vCAC uses vCenter as an endpoint to allocate resources to different enterprises within the business. Complete these tasks to configure the Production Resource vCenter as an endpoint in vCAC:

1. Log in to the vCAC web console and start configuration. Open a web browser and navigate to https://vcacwebserverhostname.domain.com/vcac as the vCAC administrator. Refer to the vCAC 5.2 Installation Guide for more information.

2. Create one new resource vCenter Credential under vCAC Administrator-Credentials. Create a username and password to connect to your resource vCenter Server.

3. Create the resource vSphere (vCenter) Endpoints under vCAC Administrator-Endpoints.

4. Create a virtual machine to be used for the vCAC Proxy Agent Machine. For more information, refer to the vSphere Virtual Machine Administration Guide.

5. Install Windows Server, the vCAC Proxy Agent Machine guest operating system, on the vCAC Proxy Agent Machine virtual machine.


DEM worker server

vCloud Suite integration (configuring endpoint)

http://www.vmware.com/pdf/vcac-52-support-matrix.pdf

https://vcacwebserverhostname.domain.com/vcac





8. Install the vSphere Proxy Agent and provide the Endpoint name created in Step 3.

Deploying vCenter Operations Manager

Complete these tasks to deploy the vCenter Operations Manager:

1. Download the OVA file for vCenter Operations Manager from VMware’s distribution website. This vApp installs two virtual machines: UI VM and Analytics VM. For more information refer to VMware Downloads.

2. Refer to the vCenter Operations Manager vApp requirements to understand the resources required for the vApp based on the scope of the resources that will be monitored. For more information, refer to the vCenter Operations Manager vApp Deployment and Configuration Guide.

3. Use the IP pools configuration option through vSphere Client to assign the network base address, netmask, and default gateway that will be assigned to a network used by the vApp.

4. Deploy the vCenter Operations Manager vApp from the vSphere Client. Set the ESXi host time, define the vCenter to be monitored, and assign a license to the vApp.

5. Configure the Chargeback Manager database to allow the maximum number of connections from the different components used by Chargeback Manager.

6. Verify the install through the vSphere Client and through a browser.

vCenter Operations Manager (vCOps) enables you to manage vCenter objects and resources. The Management vCenter and Production vCenter are added to vCOps.

To add a vCenter Server instance to vCOps, refer to the vCenter Operations Manager vApp Deployment and Configuration Guide.

Integrating vCenter Operations with the EMC Storage Analytics Suite enables full end-to-end visibility of your entire infrastructure from virtual machine to LUN and every point inbetween. This enables IT administrators to quickly visualize the health of the EMC arrays (both block and file) using a simple Performance-at-a-glance tab.

vCloud Suite Integration

EMC Integration



Figure 23. EMC Storage Analytics with VMware vCenter Operations Manager

EMC Storage Analytics (ESA) links VMware vCenter Operations for storage with EMC Adapter for VNX. vCenter Operations Manager displays performance and capacity metrics from EMC storage systems with data that the adapter provides by:

Connecting to and collecting data from block and file systems.

Converting the data into a format that vCenter Operations can process.

Passing the data to the vCenter Operations collector.

vCenter Operations Manager presents the aggregated data through alerts, dashboards, and in predefined reports that end users can easily interpret. EMC Adapter is installed with the vCenter Operations administrative user interface.

To install and configure EMC Storage Analytics with VMware vCenter Operations Manager refer to the EMC Storage Analytics for VNX Installation and User Guide.

Solution Verification


Chapter 5 Solution Verification


Overview .................................................................................................................. 53

Verifying vSphere implementation ........................................................................... 53

vCNS implementation ............................................................................................... 57

Verifying vCAC implementation ................................................................................ 58

vCenter Operations Management ............................................................................. 78



Overview

This chapter shows how to verify that the vCloud components were implemented correctly and that the VSPEX private cloud can be managed successfully to deliver infrastructure as a service.

Verifying vSphere implementation

This section shows how to verify the implementation of vSphere and integration software.

Perform these tasks to prepare the production resource vCenter so that it can be used as an endpoint in vCAC. This section assumes that the second vCenter Server instance has already been installed according to the steps detailed in Chapter 4.

1. Create a virtual datacenter for the production resource cluster. For more information, refer to the vCenter Server and Host Management Guide.

2. Type the vSphere license keys in the vCenter licensing menu. For more information, refer to the vSphere Installation and Setup Guide.

3. Connect the resource vCenter to the resource cluster ESXi hosts. For more information, refer to the vCenter Server and Host Management Guide.

4. Create a different resource cluster for the tenant departments (such as finance and manufacture) and move the correct ESXi hosts into them. For more information, refer to the vSphere Resource Management Guide.

5. Perform ESXi host discovery from the Unisphere console. For more information, refer to Using EMC VNX Storage with VMware vSphere–TechBook.

Through the VSI plug-in LUNs can be created on storage pools on the VNX and presented to clusters within vCenter. This section assumes that a storage pool designed according to VSPEX private cloud recommendations already exists on the VNX.

Create and assign LUNs with EMC VSI

After the storage pool has been created, configure, mask, and format the LUNs as VMFS datastores through the EMC VSI:

1. Right-click the vSphere cluster to which the new storage will be assigned, and select EMC > Storage Manager > Provision Storage, as shown in Figure 24.

Preparing production vCenter resources

VSI operations



Figure 24. Launch EMC VSI storage provisioning wizard from vSphere Client

2. From this point, the vSphere administrator is guided through a process which will result in a newly created VMFS datastore being mounted on all ESXi hosts in the vSphere cluster with access to the storage array.

For more information on provisioning storage through the VSI plug-in refer to EMC VSI for VMware vSphere: Unified Storage Management Product Guide Version 5.6.

3. When you click Finish, the EMC VSI wizard:

Configures the LUN.

Masks the LUN to the vSphere cluster on the array.

Rescans the ESXi server HBAs.

Formats and mounts the datastore to the ESXi servers in vSphere.

Figure 25 shows the tasks that are completed.



Figure 25. Summary of vSphere tasks executed by EMC VSI when provisioning storage

As a result, the new VMFS datastore is mounted and immediately available for use across all ESXi members of the vSphere cluster, as shown in Figure 26.

Figure 26. 200 GB VMFS datastore ‘EPC_Prod2_Fin_LUN11’ available as a storage resource

This solution uses EMC PowerPath/VE on all ESXi servers for multipathing for all SAN storage. PowerPath/VE claims the new device once it is mounted to the ESXi server, therefore no user interaction is required.

To access path management details for a storage device, right-click a datastore and select Manage Paths, as shown in Figure 27.

Note: Alternatively, in the Device List under the Storage Adapters, where the device owner is displayed as PowerPath, right-click and select Manage Paths.



Figure 27. Path management details for newly created 200 GB LUN/datastore

All of the available paths for the device managed by PowerPath/VE show as Active. As shown in Figure 27, this device has four active paths because each of the two HBAs on this ESXi server is zoned to a single VNX array port on SPA and SPB. PowerPath/VE optimizes load balancing of I/O across these paths, nondisruptively manages any path failures, and automatically restores any failed paths when made available again.

This process for storage provisioning with the EMC VSI should be repeated for each LUN required. For this solution, we created five similar-sized datastores from the Prod-2 storage service level storage pool for the Finance Organization.

Once all five datastores have been provisioned successfully, the next step is to create a datastore cluster in vSphere and place the five datastores into it. A datastore cluster is created by using the new Datastore Cluster wizard in vCenter.

When configuring the datastore cluster, enable the Storage DRS feature with the Fully Automated option. Retain the default settings in the SDRS Runtime Rules, as this solution does not use the I/O metric-based load balancing because storage is FAST-enabled on the array.

For instructions and guidance on configuring and creating datastore clusters in vCenter, refer to VMware vSphere 5 Documentation Center: Creating a Datastore Cluster.

Once the wizard has completed, the datastore cluster is fully configured in vSphere, as shown in Figure 28.

Configure datastore clusters in vSphere



Figure 28. Datastore cluster fully configured in vSphere

Verifying vCNS implementation

vCNS VXLAN enables technology for network virtualization, providing network abstraction, elasticity, and scale across the datacenter without any physical network reconfiguration.

Multiple virtual wires or logical networks can be created using VXLAN to provide isolation for various tenants or provisioning groups within the business.

Complete these tasks to create a VXLAN to separate tenant networks:

1. Prepare the network for VXLAN virtual wires. Ensure the current software and hardware environment meets the prerequisites for creating VXLAN, the prerequisites can be checked in VMware vCloud Networking and Security Documentation.

2. Associate resource clusters with distributed switches. Log in to the vSphere Client and map each resource cluster to the distributed switch.

3. Assign the segment ID pool and multicast address range. In the vSphere Client, assign the segment ID and multicast address range to vShield Manager under Inventory -> Hosts & Clusters, select a datacenter resource, and under Network Virtualization -> Preparation -> Segment ID.

4. In the vSphere Client, add a network scope under Inventory -> Hosts & Clusters -> Network Virtualization -> Network Scope.

Separating tenant networks





5. In the vSphere Client, add a VXLAN virtual wire under Inventory -> Hosts & Clusters -> Network Virtualization -> Network. This step adds a separate VXLAN virtual wire for provisioning groups.

After creating virtual wires or logical networks for provisioning groups, vCNS license activated components such as vCNS Edge gateway and vCNS App can be used to provide networking services and security within the business. For more information on configuring VXLAN and edge gateway services, refer to the VMware VXLAN Deployment Guide.

Verifying vCAC implementation

vCAC provides the ability to divide a shared infrastructure into logical units and logical capacities. These can, in turn, be handed over to different business units where users can choose from a self-service catalog of customized virtual machine blueprints.

The resource clusters supporting the various organizations within the enterprise are managed by a vCenter server located in the management cluster. This vCenter server operates as the vSphere end-point for VMware vCAC, to which all of the underlying vSphere resources are made available.

This section explains how to:

Create enterprise groups and Provisioning groups within vCAC to enable multitenancy by allocating and reserving resources through the vSphere endpoint configured in Chapter 4.

Set up vCAC blueprints and making them available to groups within the enterprise.

Create and request virtual machines through blueprints.

Reconfigure and decommission virtual machines through a self-service portal.

vCAC is built to work with existing infrastructures. It supports the differing requirements of an enterprise’s many business units and integrates with a wide variety of existing IT systems and best practices.

A number of user roles and responsibilities are defined and used in the structure of vCAC. The administration of users and compute resources in vCAC is managed through the vCAC console, which is the administrative portal. The relevant groups, users, and roles are summarized in the following sections:

Cloud administrator

Enterprise groups

Provisioning groups

Cloud roles and resource administration

http://www.vmware.com/files/pdf/techpaper/VMware-VXLAN-Deployment-Guide.pdf

http://www.vmware.com/files/pdf/techpaper/VMware-VXLAN-Deployment-Guide.pdf



Cloud administrator

The cloud administrator role is responsible for configuring resource endpoints and enterprise groups, where enterprise group administrators and their respective compute resources need to be defined.

A primary task of the cloud administrator is to configure the endpoints which vCAC then uses for provisioning compute resources and operations. In this solution, the Production vCenter was added as a vSphere endpoint called vSphere (vCenter) which is used by vCAC for compute resources.

The configuration of the endpoint is displayed in Figure 29.

Figure 29. Endpoint configurations for vCAC

The cloud administrator is responsible for creating enterprise groups and assigning compute resources to them. Figure 30 highlights the compute resources of the Manufacturing cluster in vSphere being selected to support the WayneEnterpriseMfg group.



For more information on creating enterprise groups and cloud administrator operations, refer to the vCloud Automation Center Operation Guide.

Figure 30. Enterprise groups created and managed by vCAC cloud administrator

Figure 31. Enterprise groups resource defined vCAC cloud administrator

Enterprise groups

Enterprise groups contain enterprise administrators who manage their respective cloud resources, as defined by the cloud administrator. Enterprise administrators are



responsible for configuring and creating provisioning groups for individual departments and resource reservations and customizations for network, storage, compute and cost profiles. Distributed virtual port groups created on the vSphere distributed switch are available as network paths to be selected when creating a reservation. Logical networks created through vCNS are also available as network paths to be selected. Approval groups and policies are also defined by the enterprise administrator.

The enterprise administrators can create a vSphere vCenter virtual reservation, which is a share of one virtualization compute resource’s resources dedicated to a particular provisioning group for use in provisioning virtual machines. The reservation policies can be used to classify the reservation.

Figure 32. Storage resource reservations as managed by the vCAC enterprise administrator

The storage administrator can create different kinds of storage pools based on different arrays, like a ‘gold’ class based on a VSPEX storage pool and a ‘silver’ class consisting of all SAS drives in a RAID5 configuration (refer to Appendix A). Then the enterprise administrator can tag them in different reservations, as shown in Figure 33.

Creating vSphere virtual resource reservation and cost profile



Figure 33. Assigning cost profiles

Figure 34 displays the configured provisioning groups for the Finance and Manufacturing departments. The provisioning group WayneEntMfgProvGrp1 is highlighted, and contains provisioning group managers responsible for the Manufacturing department.

Figure 34. Provisioning groups managed by vCAC enterprise administrator

A cloud or enterprise administrator can edit this provisioning group to define the group manager, support, and user roles for the Manufacturing department.

The enterprise administrator has the responsibility of assigning compute, storage, and network resources and making them available to the provisioning group manager. In this solution the resources are carved from the cluster in vSphere allocated to this enterprise. Resources can be reserved by the enterprise



administrator for their provisioning groups using reservation policies. These policies cover reservations across memory, storage, and network resources.

For more information on creating provisioning groups, reserving resources, and reservation policies, refer to the vCloud Automation Center Operation Guide.

Provisioning groups

The users in these provisioning groups are the users and consumers of the infrastructure provided to them by their enterprise administrator:

Provisioning Group Manager (PGM)–A PGM can access all virtual machines, create and publish blueprints for end users, manage approval requests, and work on behalf of other users in their group.

Support user role–This is a helpdesk user whose role enables them to work on behalf of other group users where required for troubleshooting and support.

User role–These users are the end users in the context of vCAC, who can deploy from the blueprints made available to them by the PGM.

Those users in the user role are the primary consumers of the vCAC self-service user portal from where they can provision and manage their virtual machines. The deployment of blueprints may be subject to approval by the PGM. This approval policy is set on a per-blueprint basis by the PGM, as highlighted in Figure 35.

For more information on roles within a provisioning group refer to the vCloud Automation Center Operation Guide.

Figure 35. Blueprint approval policy as set by provisioning group manager

This solution environment, as an example, is based around a fictional company named Wayne Enterprises, which contains two production organizations: Manufacturing and Finance. These organizations are supported by users in the IT department who are directly responsible for supporting the systems and applications of each organization.



A blueprint is the complete specification for a virtual machine, determining the machine’s attributes, the manner in which it is provisioned, and its policy and management settings. When a user requests a machine through the self-service portal, they must select the blueprint from which it will be created. Blueprints specify the workflow used to provision a machine and define the build profiles containing additional provisioning information as needed. The blueprint also sets the policies that apply to a machine, such as any approvals required, expiration date, and owner operations.

When a blueprint is created, information is specified using controls in the console and custom properties. A local blueprint is created by a PGM for a single provisioning group and is available only to that group’s members. A global blueprint is created by an enterprise administrator or PGM and can be selected for inclusion among a provisioning group’s local blueprints by the PGM.

Blueprints can be single or multi-machine, including multi-tier enterprise applications which require multiple components (application, database, or web). A multi-machine blueprint contains multiple individual machine blueprints.

Creating blueprints

Enterprise administrators and PGMs (or a user who is both) can create blueprints. A user who is both an enterprise administrator and a PGM can choose whether to create a global blueprint, which is available to all provisioning groups, or a local blueprint which is confined to a single group.

To create a local blueprint:

1. Log in to the vCAC console as a PGM.

2. Select Provisioning Group Manager -> Blueprints in the activity pane

3. Click New Blueprint and select Virtual, as shown in Figure 36.

Figure 36. Create new VMware vSphere virtual machine blueprint

You can create a blueprint from the beginning or copy it from an existing master blueprint. Copying an existing master blueprint may be useful or more efficient in circumstances where the new virtual machine shares a large number of identical properties, meaning less customization is required in building the blueprint.

vCAC blueprints



Blueprint creation involves completing information across four separate tabs:

Blueprint Information Build Information Properties Security

Each of the four tabs contains several input fields which combine to define the final product of the blueprint.

Blueprint Information

The new blueprint is set as a master blueprint which then allows it to be copied for subsequent blueprint creation. Other properties such as reservation policy, provisioning group and virtual machine name prefix, and machine daily cost are configured here, as well as specifying that this blueprint requires the approval of the provisioning group manager when deployed by users. The process of creating a new blueprint is shown in Figure 37.

Figure 37. Create new blueprint

Build Information

In the next tab, Build Information, the blueprint type is set for Server and the Clone action is selected which means that a cloning workflow will take place, based on an existing template in vSphere. In this example, a number of standard Windows and Linux virtual machine templates are pre-built in vSphere. Select the desired template by clicking next to the Clone from text field, as highlighted in Figure 38. The Windows2008R2SP1 template is selected for this example.



Figure 38. Select vSphere template to clone from

The virtual machine resources are, at a minimum, set to how the template is configured. The blueprint can be set to allow these values to be increased at a later stage up to a maximum value. Figure 39 shows where to set the machine resources maximums and where to specify the storage service level (Storage Reservation Policy).

Figure 39. Complete build information for new blueprint



For this example, the minimum machine resources are inherited from the vSphere template and the maximum values have been manually set at double the minimum values. This means that at deployment time, this virtual machine can be deployed with up to the maximum values. Later in this configuration, we provide cloud users with the ability to reconfigure the virtual machine resources after deployment.

Properties

The Properties tab contains a selection of build profiles. Some are specific to each operating system, while others are applicable to all virtual machines. A build profile contains a set of properties to be applied to a machine when it is provisioned. For this blueprint, we selected a build profile specific to Windows Server 2008 and Windows 2012 virtual machines. The selection process is shown in Figure 40.

Figure 40. Select custom build profiles for virtual machine

Security

Finally, for this blueprint creation, in the Security tab we specify which users can access this blueprint, and specify machine operations and reconfiguration options, as shown in Figure 41.



Figure 41. Enable virtual machine operations for cloud user

For more details on creating virtual machine blueprints refer to the vCloud Automation Center Operating Guide v5.2.

After the blueprint has been created, it is then available for cloud users to deploy from the user self-service portal, as shown in Figure 42.

Figure 42. Deploy new virtual machine blueprint from self-service portal

The PGM enables a cloud user to reconfigure one of their virtual machines in the blueprint. A machine owner can make any of the following changes to a provisioned machine:

Increase or decrease memory

Deploy virtual machines from blueprints

Reconfigure existing virtual machine



Increase or decrease the number of CPUs

Modify storage by adding, removing, or increasing the size of volumes (SCSI disks only)

Modify networks by adding, removing, or updating network adapters

Changes to each of these parameters are subject to the limits defined in the blueprint that was used to provision the machine originally. Figure 43 highlights an example of a virtual machine owner reconfiguring the storage resources of a virtual machine from the self-service portal.

Figure 43. Reconfigure virtual machine resources from self-service portal

Reconfiguration requests can also be made subject to approval, as with provisioning requests. When the request is approved, the reconfiguration is ready to be executed. This is shown in Figure 44.

Figure 44. Set execution options for virtual machine reconfiguration



The three options for when to execute the reconfiguration are:

Immediate —The reconfiguration is automatically queued for execution after all required approvals have been given.

Scheduled —The machine owner specifies a date and time to execute the reconfiguration. Execution can be rescheduled if the request is not approved until after the scheduled time.

Queue for Owner —The reconfiguration must be triggered manually by the machine owner after all approvals have been given.

When the machine’s lease expires, or the machine is manually expired, the machine is either archived or destroyed, depending on whether its blueprint specifies an archive period. When the end of the archive period is reached, or if there is no archive period, the virtual machine is destroyed, as shown in Figure 45.

Figure 45. Decommissioning a virtual machine

This private cloud environment requires flexible metering and costing models that can account for the utilization of all resources within the environment. VMware vCAC contains a built-in cost reporting solution for all the physical and virtual resources allocated and consumed within the private cloud. Managing the lifecycle of cloud resources by reclaiming unused, underutilized, or expired resources is also measured and captured by the vCAC reporting portal, along with estimated savings associated with the reclaimed resources.

Note: The scope of this reporting functionality is limited to the resources consumed within VMware vCAC.

The vCAC reporting portal is available to the following users:

Support user

Provisioning group manager

Enterprise administrator

vCAC administrator

Decommission a virtual machine

Chargeback and metering



Any of these users can access the vCAC reports web site by clicking Reports in the activity pane.

Report types

The first available, or top-most, report is the Executive Summary report (shown in Figure 46) which provides an overview of provisioning and resources for the entire vCAC site, by machine types of virtual, physical, and cloud. The scope of this solution includes virtual resources; therefore all reports in this section reflect this.

Figure 46. Executive Summary Virtual for vCAC virtual machines

Included in this Executive Summary report, as shown in Figure 46, are graphics representing the distribution of virtual machines by blueprint, group, state, and reservation. The table provides totals for physical, reserved, allocated and used virtualization resources.

More general capacity usage reports are also available in the vCAC reporting portal. The capacity usage reports, as displayed in Figure 47, provide breakdowns of machine quota, memory, and storage usage by provisioning group, virtualization compute resource, machine owner, and blueprint. Only provisioning groups with reservations of the selected type are displayed.



Figure 47. Manage compute resource cost and storage cost profiles

Figure 47 displays the capacity usage, filtered by group, for both the Finance and Manufacturing organizations used in this solution. This report shows resources reserved, allocated, and used by all provisioning groups. The summary table is broken into three columns containing capacity usage information for virtual machine, memory, and storage.

Cost profiles

To enable calculation of a machine’s cost for display to machine owners, requesters and administrators, and for chargeback reporting, you can associate each virtualization compute resource with a cost profile. Additionally, for finer definition of storage cost for virtual machines, you can also associate each known vCAC storage path (datastore/LUN) on a compute resource with a storage cost profile.

A daily cost can also be specified in each blueprint. Cost profiles and daily blueprint cost are optional. Cost profiles are assigned to compute resources by enterprise administrators. Blueprint costs are specified by enterprise administrators for global blueprints and PGMs for local blueprints, as shown in Figure 48.

Figure 48. Manage compute resource cost and storage cost profiles



For this solution, a storage cost profile has been created to represent each of the storage service levels presented and available within VMware vCAC. Figure 49 displays all cost profiles in this solution.

Enterprise administrators can assign or change a compute resource’s cost profile using the Edit compute resources page. Each cost profile specifies these costs:

Compute resource cost–A daily cost per GB of memory capacity specified in the virtual blueprint or installed in the physical machine

CPU cost–A daily cost per CPU specified in the virtual blueprint or installed in the physical machine

Storage cost–A daily cost per GB of storage capacity as specified in the virtual blueprint

Storage cost profile A storage cost profile, highlighted in red in Figure 49, contains only a daily cost per GB of storage. If you assign a storage cost profile to a datastore or LUN, this storage cost overrides the storage cost in the cost profile, highlighted in blue in Figure 49, assigned to the compute resource, if any. These daily costs are added to that specified in the blueprint, if any to be displayed and charged back to machine owners and provisioning groups. Also included in the Configuration tab for the vCAC compute resources is the Storage Reservation Policy. This policy has been aligned with the storage cost profiles for the associated datastore clusters.

Figure 49. Edit vCAC compute resource cost and storage cost profiles

Chargeback reports Taken together, the combined costs enable the generation of daily costs per machine, and chargeback reports totaling these costs for each machine and for all machines on



each reservation over a specified period. Once the appropriate cost profiles have been created and assigned to the relevant resources in vCAC, then the chargeback section of the vCAC reporting portal provides access to a number of chargeback reports.

From the vCAC reporting portal, chargeback reports can be separated and calculated by machine type, physical or virtual, and can be viewed as follows:

Chargeback by Group by Allocated Resources

Chargeback by Group by Reservation

Chargeback by Owner by Allocated Resources

The reports can be filtered by date, provisioning group, and machine type.

Chargeback by Group by Allocated Resources: This report (shown in Figure 50) shows the cost for all provisioning groups or a selected provisioning group of all machines or all machines of a selected type owned by members of the group, over a specified period of time. For example, you can see the cost of virtual machines owned by a particular group during a particular week, or by all provisioning groups over a year. Costs are shown for individual machines and totaled for all machines owned by a provisioning group.

Figure 50. Chargeback by group allocated resources

The chargeback report displayed in Figure 50 is filtered by date for one month, by Provisioning Group (WayneEntMfgProvGrp2), and by Machine Type (Virtual). Included



in the output of this report are the daily charges, the number of days charged for each, their total cost, and the combined cost of all machines for the provisioning group.

Chargeback by Owner by Allocated Resources: This report shows the cost for all machines’ owners or a selected owner of all machines or all machines of a selected type owned, over a specified period of time. For example, you can see the cost of machines of all types owned by a particular owner during a particular week, or by all owners over a year. Costs are shown for individual machines and totaled for all machines owned by a user.

The chargeback report displayed in Figure 51 contains chargeback information for two provisioning groups: WayneEntMfgProvGrp1 and WayneEntMfgProvGrp2. This report has been run by the ‘mfgIT’ user who is an enterprise administrator for the Manufacturing organization; therefore this user will only see information specific to provisioning groups in their own organization.

The total costs stated in this report are totals for the amount of compute resources reserved for these provisioning groups. These costs for reservations are irrespective of how much of those resources are actively being consumed.

Figure 51. Chargeback by group by reservation

Chargeback by Owner by Allocated Resources: This report shows the cost for all machines owners or a selected owner of all machines or all machines of a selected type owned, over a specified period of time. For example, you can see the cost of virtual machines owned by a particular owner during a particular week, or by all owners over a year.



Figure 52. Chargeback by owner by allocated resources

The chargeback report in Figure 52 is filtered to report on the total cost of virtual machines owned by an application orientated owner in the manufacturing organization named mfgAppOwner. The time period for this report is for the past month.

The output of this report contains the daily charges per individual virtual machine, the number of days charged for each virtual machine, their total cost, and the combined cost of all machines for that owner.

Reclamation reports Reclamation reports provide information on all machines or machines of a particular type that have been reclaimed (destroyed or decommissioned) during a particular period. This includes machines destroyed or decommissioned after lease expiration or manual expiration or machines manually destroyed or decommissioned, as well as virtual machines destroyed by a reclamation workflow initiated by an enterprise administrator.



Figure 53. Dashboard for reclaimed virtual machines in the enterprise private cloud

The Reclaimed Resources Dashboard report, as displayed in Figure 54, totals reclaimed machines, or machines of a selected type, during the last 7, 30, 90, and 365 days.

Further detail can be gained with regard to who the reclaimed virtual machine owners were and what the operational savings were due to the reclamation.

Figure 54. Reclamation Savings by Owner report

The Reclamation Savings by Owner report, as displayed in Figure 54, shows information for each owner of one or more machines reclaimed since the specified start date. The savings quoted in this report are based on the difference between the hypothetical cost of all machines owned by the provisioning group since the specified start date if no machines had been reclaimed, and the actual cost including machine reclamation.



vCenter Operations Management

Managing operations in cloud environments includes ensuring service levels, problem solving, troubleshooting, and preventative maintenance. Successfully meeting these requirements involves analysis of current utilization, forecasting of future needs, and the optimization of current resource consumption by reclaiming underutilized resources.

The in-depth infrastructure and operations information provided by VMware vCenter Operations Management (vCOps) eliminates time-consuming problem resolution processes through automated root cause analysis. Self-learning performance analytics and dynamic thresholds adapt to the environment to simplify operations management and eliminate false alerts. Integrated smart alerts for health, performance, and capacity degradation identify performance problems before they affect end users. Advanced capacity analytics allow administrators to optimize virtual machine density and identify capacity shortfalls before they affect end users.

The VMware vCenter Operations Management Suite Enterprise Edition provides flexibility with advanced features that extend monitoring, analytics, and reporting capabilities and it is suitable for solutions of any size.

This VSPEX IaaS solution uses vCOps to monitor both the management layer and the compute resources used in production. VMware vCOps is not integrated with, or aware of, VMware vCAC v5.2, and therefore has no concept of what resources are owned by vCAC. However, both the Management and Production (vCAC resources) components are managed by separate vCenter servers, which means their respective resources can be managed and monitored separately in vCOps.

vCenter Operations Manager dashboard Once implemented and configured, the vSphere UI provides a comprehensive insight into the environment, as shown in Figure 55. The main dashboard is divided into three logical entities providing high-level information about the current overall health and issues of all managed resources, potential future issues and risks in the environment, and resource efficiency trends in the environment.

Figure 55. vSphere UI dashboard high level overview

Monitoring vCenter objects and resources



The three primary logical entities in vCOps can be explained as follows:

Health calculates health scores based on patented algorithms that dynamically calculate thresholds by observing the behavior trends of the cloud environment and gives visibility into the red, yellow, green status of the virtual machines, datastores, and clusters. You can see where the problems are, what the issue is, and if there is any pattern of abnormal behavior in the environment.

Risk provides insight into the resource consumption to tell you not only when you will run out of capacity but also which resources will run out.

Efficiency aims to be proactive in order to optimize your environment and reclaim waste.

vCOps badges

vCenter Operations Manager uses badges to illustrate derived metrics to provide an overview of the state of the virtual environment or an individual object. These badges serve as focus points to narrow the scope of a potential problem and provide details about the cause of the problem.

Major and minor badges are color coded and range from a healthy green to a potentially problematic yellow, orange, or red. Badges are organized in a simple hierarchy in which the scores of minor badges contribute to the scores of major badges.

Scores might reflect a healthy state or a potential problem depending on the type of badge. For example, low scores for health, time remaining, and capacity remaining might indicate potential problems, while low scores for faults, stress, or anomalies indicate a normal state.

Navigation and troubleshooting

The vCOps dashboards provide a wide number of different insights into infrastructure condition and enable simple and easy analysis of the root cause of problems.

For example, while the overall environment condition for health is represented with a green badge value of 97, by hovering the mouse over one of red marked areas on the Health Weather Map we get brief information about a fault. In the example used in Figure 56, the red icon on the bottom row of the Health Weather Map is related to a particular VNX data store, indicated with a Health value of 0 and 100 value for Faults. Note that there is only a single fault, but the 100 value represents a percentage faulted.



Figure 56. Health dashboard showing immediate issues in environment

Double click this red area to display further details about the issue. vCOps automatically navigates the view to the source of the issue, which in this example is VNX-based datastore, VNX-VSI-DS02. Select the Faults badge to show more comprehensive details which indicate lost storage connectivity. Event details including fault criticality, resource type, resource name as well as datastore workload, space and performance related metrics are available, as shown in Figure 57.



Figure 57. End to end fault analysis covering actual fault, causes, and solution details

The level of troubleshooting supporting the monitoring differentiates vCOps from standard monitoring solutions. Instead of providing large amount of alerts when certain thresholds are reached or abnormal states occur, vCOps provides detailed information on the actual issue, and its potential causes and solutions, as shown in Figure 57. Selecting What does that mean? from the dashboard provides a detailed description of root cause and a step-by-step procedure with links to appropriate knowledge base articles you should follow to determine the current issues in your environment. An example of this solution detail is displayed in Figure 58.

Figure 58. VMware vCOps root cause analysis solution details



VMware vCOps integration with EMC Storage Analytics (ESA) software combines the features and functionality of VMware vCenter Operations Manager with VNX. It delivers custom analytics and visualizations that provide deep visibility into your EMC infrastructure and enable you to troubleshoot, identify, and take quick action on storage performance and capacity management problems.

Within the vCOps custom portal, ESA presents three separate dashboards: two are universal and one is specific to VNX. Each dashboard is fully customizable and can be adjusted to display the required details and metrics or additional widgets.

By default, ESA enables the following dashboards in the vCOps custom portal which provide information about VNX storage systems:

EMC Storage Topology dashboard – Provides view of resources and relationships between storage and virtual infrastructure objects, as shown in Figure 59.

EMC VNX Storage Metrics dashboard – Displays resource and metrics for storage systems.

EMC VNX Overview dashboard – Represents a single view of performance and capacity of VNX resources.

EMC Storage Topology dashboard

The EMC Storage Topology dashboard is shown in Figure 59. It shows the VNX arrays that are configured – these are referred to as adapter instances. When an array is selected from the Storage System Selector pane, the Topology and Resources panes are populated with associated underlying components. ESA pulls in all related VMware objects, making it possible to navigate end-to-end into the underlying storage array components – from vSphere datastore cluster and LUN to storage pools and storage processor. The instance in Figure 59 shows that EPC_Prod2_LUN10 has datastore EPC_Prod2_LUN10 built on it. The LUN is created from the vCAC pool and uses Storage Processor A. These details are presented graphically in the Storage Topology and Health pane and also presented as parent and child resources in their respective panes.

Monitoring VNX health and resources



Figure 59. EMC VNX Storage Topology dashboard

The second universal dashboard, shown in Figure 60, is for EMC Storage Metrics. Each EMC resource has a set of metrics associated with it and is displayed in a graph. Navigation is driven from the top down, so once the storage system and specific resource have been chosen, the user can select multiple metrics to be displayed in the Metric graph.



Figure 60. ESA – EMC Storage Metrics displaying VNX LUN metrics

VNX arrays have their own specific dashboard, EMC Performance Overview, which presents details in the form of heat maps, as shown in Figure 61. This dashboard covers the main storage system resource types – thin pools, storage groups, LUNs, storage processors (VNX), FAST Cache Performance (VNX) – and provides a few metrics from each one. The heat map colors work on two different levels: there is a green-to-red legend for some which represents either usage (for example, thin pool allocation) or performance (for example, latency) and there is a blue legend for relative usage across that metric within an array (for example, total writes). For any one of the objects shown, a full historical perspective is available on the EMC Storage Metrics page.



Figure 61. ESA – VNX Overview dashboard displayed object heat maps

VMware vCOps provides powerful capacity planning functionalities which help predict behavior and evaluate the potential impact of future growth on the underlying resources supporting the enterprise private cloud environment.

The capacity planning component of vCOps provides statistics on the current utilization, as shown in Figure 62, but can also provide a prediction for a “What-If scenario” where the infrastructure environment may be influenced by an increased or decreased number of ESXi hosts, storage, or virtual machines on existing or new consumption profiles. By implementing the scenario, vCOps will model the predicted impact so that you know in advance what your capacity requirements will be under such circumstances.

Capacity planning and optimization



Figure 62. Virtual Machine Capacity details for management cluster

As shown in Figure 62, the capacity-remaining figures are based on demand and consumption trends of the currently operating virtual machines.

To plan for capacity requirements for future growth, you can create a What-If scenario which contains a virtual machine profile that can be based on an existing virtual machine or specified manually, as shown in Figure 63.

Figure 63. Specify virtual machine configuration for What-If scenario

As shown in Figure 63, the virtual machine profile can be tailored to specify not just the allocation of resources but also their actual usage and consumption. Once this scenario of adding ten new virtual machines has been successfully run through, the details displayed for virtual machine capacity are updated to display the current actual capacity remaining as well as with the What-If scenario implemented, as shown in Figure 64.



Figure 64. What-If scenario – Adding ten new virtual machines

Under- or over-utilized virtual machine dashboard In situations where resources are limited, vCOps is capable of identifying reclaimable, underutilized resources in idle or oversized virtual machines. The resource-optimization-related waste dashboard will compare configured and recommended CPU and memory metrics and determine oversized virtual machines according to actual resource consumption over defined period of time.

The definition of under- or over-utilized virtual machines can be customized, is policy based, and can be edited to suit specific business requirements, as shown in Figure 65. Multiple policies can be created and applied as appropriate.

Figure 65. Edit policy to specify thresholds for underutilization of virtual machines



The dashboard detail provides a list of oversized virtual machines and recommendations of appropriate values for CPU and RAM resources according to real consumption history, as shown in Figure 66.

Figure 66. List of oversized virtual machines with suggested optimal resource configurations

vCOps reports The Reports tab in vCOps provides a formal reporting structure around the various views and summaries available in the Planning section. Each report can have a specific schedule attached or can be executed manually, as displayed in Figure 67 for the Virtual Machine Capacity Overview Report.



Figure 67. Schedule a report in vCOps or run the report manually

Note: The New Report Schedule functionality is password protected and provides the ability to email the completed report to multiple addresses.

Once a report has successfully completed, it is available, along with previously run instances of the same report, to be downloaded in either PDF or CSV format, as shown in Figure 68.

Figure 68. Successfully completed reports available for download

Resource Sizing Guide


Chapter 6 Resource Sizing Guide


Overview .................................................................................................................. 91

Cloud management environment sizing ................................................................... 91

Summary .................................................................................................................. 93



Overview

Understanding the performance requirements of a private cloud environment can be a challenge. The EMC VSPEX solution for IaaS simplifies this process significantly because the solution is based on an existing VSPEX private cloud platform with known levels of performance and scalability.

This chapter provides sizing information for the additional management and orchestration components.

For this sizing guide there are two distinct elements:

The sizing of the management environment components that form the private cloud

The supporting infrastructure components on which the virtualized workloads will run

Cloud management environment sizing

The other documents referenced in this section provide guidelines for configuring the characteristics of each component in a solution. Table 4 provides examples of how the solution is sized according to those guidelines.

For more details, refer to EMC VSPEX Private Cloud VMware vSphere 5.5 for up to 1,000 Virtual Machines Proven Infrastructure Guide.

As mentioned in the VSPEX Private Cloud document, a reference virtual machine has 1 vCPU, 2 GB RAM, and 100 GB storage space.

Table 4. Management vCenter requirements

Server role vCenter components Recommended hardware specifications

Number of VSPEX reference virtual machines

SQL Server database server (SQL-DB01)

vCenter database CPU: 2.4 GHz 4-core or equivalent RAM: 8 GB

4

Disk: 40 GB

Network: 1 Gb/s

vCenter Server (vCenter-01)

SSO vCPU 2 RAM: 10 GB 5

Inventory Disk: 100 GB

vCenter Server Network: 1 Gb/s



Table 5. vCAC Up to 1,000 machines (physical or virtual)

Server role vCAC components Recommended hardware specifications


Database server vCAC database CPU: 2.4 GHz 4-core or equivalent RAM: 8 GB

4

Disk: 40 GB

Network: 1 Gb/s

vCAC server/web server

Manager Service DEM Orchestrator

CPU: 2.4 GHz 4-core or equivalent RAM: 4 GB

4 (plus 4 for failover server)

Model Manager (web and data) portal website

Disk: 40 GB

Reports website Network: 1 Gb/s

DEM machines Distributed Execution Managers (DEM) Workers (one or more)


4

Disk: 40 GB

Network: 1 Gb/s

Agent machines vCAC agents (one or more)


4

Disk: 40 GB

Network: 1 Gb/s

Note: There is one active web server, with one server for failover. In this configuration, the Manager Service can be co-hosted with the web components.

Table 6. vCOps hardware requirements

Server role vCOps components Recommended hardware specifications


Management UI VM CPU: 2.4 GHz x 2 vCPU or equivalent RAM: 7 GB

4

Disk: 100 GB



Server role vCOps components Recommended hardware specifications


Network: 1 Gb/s

Analytics Analytics VM CPU: 2.4 GHz x2 vCPU RAM: 9 GB

8

Disk: 800 GB

Network: 1 Gb/s

Table 7. Production vCenter requirements

Server role vCenter components Recommended hardware specifications


SQL Server database server (SQL-DB02)

vCenter database CPU: 2.4 GHz 4-core or equivalent

RAM: 16 GB

8

Disk: 40 GB

Network: 1 Gb/s

vCenter Server (vCenter-02)

SSO vCPU 2 RAM: 10 GB 5

Inventory Disk: 100 GB

vCenter Server Network: 1 Gb/s


vCNS Manager CPU: 2.4 GHz 2-core or equivalent

RAM: 8 GB

Disk: 60 GB

Network: 1 Gb/s

4

This would require a total resource requirement of 57 VSPEX reference virtual machines.

Summary

The EMC VSPEX solution for IaaS creates a robust, scalable platform infrastructure that includes the best features of the underlying EMC and VMware technologies and layers them into the complete solution. Because it is built on the VSPEX private cloud



platform, sizing the appropriate solution becomes significantly easier since EMC has provided solutions with known sizing metrics.

This solution provides several recommendations and examples for enabling IaaS within the business. VSPEX partners can use these recommendations to create different types of resource pools or provisioning groups to meet specific application and performance requirements of enterprises within a customer environment. This solution uses the same design, hardware, and configuration described in the VSPEX Proven Infrastructure as a base architecture to provide IaaS within a business.

It discusses two additional types of storage service levels in Appendix A on page 96 to give customers the flexibility of deploying virtual machines based on performance and capacity requirements. These additional storage levels should have their own set of disks separate from the disks required in the VSPEX Proven Infrastructure.

However, EMC recommends that you use only the VSPEX storage layout to calculate and support the exact number of VSPEX reference virtual machines required. Deploying virtual machines on different storage service levels can reduce the total number of supported reference virtual machines.

When building the cloud infrastructure, you should note certain recommendations, including verifying the interconnect points of IP connected devices. Where workloads and infrastructure management exist, connections with a minimum of 10 Gb with appropriately sized switching backplanes are and should be used. Failure to observe these recommendations can result in underperforming situations, such as resource scheduling problems or failure scenarios, which require the movement of workloads as quickly as possible.

The VSPEX sizing tool can assist you in quickly, easily, and properly sizing your VSPEX private cloud environment, including the required management and orchestration applications. You can then determine the proper VSPEX private cloud reference architecture for your given workloads that has been tested, sized, and proven by EMC.

Appendix A: Storage Service Levels




Overview .................................................................................................................. 96

Storage service levels .............................................................................................. 96



Overview

The VSPEX IaaS solution using VMware vCloud Suite is designed on top of an existing VSPEX Proven Infrastructure for vSphere. The Proven Infrastructure uses a specific configuration for storage pools on the VNX to support a required number of virtual machines. As a result, this solution uses a single storage service level.

This appendix provides information on using additional storage service levels in the VSPEX Private Cloud solution. However, it should be noted that the VSPEX architecture is built for a specific number of virtual servers and using different storage levels may reduce the number of virtual servers supported.

Storage service levels

The VNX storage array provides many features that allow provisioning of IaaS. Creating storage pools with specific types of drives enables creation of different storage service levels based on pool composition and drive performance.

Multiple storage service levels can be created within a VNX array to achieve different levels of performance and most effectively align resources with performance requirements.

Storage service levels can be created across a VNX with varying availability, capacity, and performance capabilities. You can weigh each service level respective to the application and operational requirements of each line of business, with some weighted in favor of capacity and others weighted for performance. The service levels provide the highest performance where workloads demand it. The performance capabilities decrease as the service levels move toward serving less performance-intensive workloads where cost is more important. Figure 69 shows the performance versus capacity requirements for the different segments of an organization used as a reference in this solution.

Figure 69. Storage service-level positioning for VSPEX management and orchestration

The Prod-2 storage service level built according to the VSPEX Private Cloud solution storage design is the only VSPEX recommended storage service level. This section provides two additional storage layout examples for virtual machines with different

Storage levels consumable by tenants

Storage service-level offerings



performance requirements within an organization. All three storage service levels are shown in Figure 70 and described in Table 8.

Figure 70. Storage service levels on VNX systems

Note: Additional drives need to be present on the VNX to support the two storage service levels. Deploying virtual machines on these storage service levels may lower the number of reference virtual machines as specified in the VSPEX configuration.

Note: FAST VP must be enabled for all service levels so that they can take advantage of storage tiering and load balancing when different drive types are added to the storage pool.

Table 8. Storage service levels example

Service level

FAST VP SSD disks SAS/FC NL-SAS/SATA

Prod-2 Yes 2 x 200 GB 45 x 600 GB SAS N/A

TestDev No N/A 27 x 600 GB SAS N/A

Archive No N/A N/A 16 x 2 TB NL-SAS

The storage service levels are allocated as follows:

Prod-2 is configured for production systems within organizations where performance is a criterion (similar to the VSPEX private cloud storage layout).

TestDev is configured for testing and development systems.

Archive is configured for archived systems or for systems where cost, but not performance, is important.



vCAC makes these storage service levels available to the appropriate enterprise groups within the business. Provisioning groups are allocated a specific storage service level based on performance needs and requirements.

Note: If the exact number of reference virtual machines, as stated in the VSPEX Proven Infrastructure document are to be deployed, then only the Prod-2 storage service level should be used. Each Prod-2 storage service level supports up to 125 virtual machines. This Proven Infrastructure document provides exact sizing details: EMC VSPEX Private Cloud VMware vSphere 5.5 for up to 1,000 Virtual Machines Proven Infrastructure Guide.

Figure 71 shows an example of how storage resources can be mapped for a particular organization.

Figure 71. Storage service-level mapping from applications to storage array

Documents

EMC VSPEX Solution for Infrastructure as a Service with VMware