Table of contents Introduction ...
10 to the Command View server; therefore the Site Recovery Manager server and the HP EVA SRA can be installed either on a physical machine or on a vi
11 Figure 3. Protected Site Array Manager configuration Figure 4. Recovery Side Array Manager configuration
12 The most common deployment scenario is to have a Command View server at Site A or Site B, managing both the protected and the recovery arrays. Fig
13 Figure 6. Active/Active Command View placement Installing EVA licenses HP StorageWorks Continuous Access EVA is needed to create replication pa
14 Figure 7. EVA Licensing in Command View and SSSU
15 Site Recovery Manager Objects relationship Site Recovery Manager introduces three management objects to handle the various elements of the configu
16 VMs in serial order because these VMs generally have dependencies between each other, and because we want the most predictability. The medium and
17 Figure 9. Site Recovery Manager datastore groups Two factors determine what causes LUNs and datastores to be grouped together and not distinctl
18 If virtual machines that have no relationship are consolidated on a single volume, those virtual machines will have to failover together. With cer
19 Figure 10. Symmetric protection groups Alternatively, Site Recovery Manager also allows asymmetric or one-to-many datastore group to DR group r
2 Introduction VMware deployments are increasingly being used for mission-critical applications. In response to this trend, VMware delivered VMware v
20 Figure 11. Asymmetric datastore group configuration Optimizing performance ESX SAN load balancing and HP StorageWorks EVA The EVA 4x, 6x, and 8
21 Figure 12. ESX multipathing default configuration The default ESX 3.5 multipath configuration may result in imbalanced SAN and EVA controller l
22 Figure 13. Optimized ESX multipathing configuration with HP StorageWorks EVA Implicit LUN transition To optimize the host data paths and reduce
23 When using Continuous Access EVA in a VMware ESX environment, HP recommends setting a preferred controller to virtual disks configured in a DR gro
24 Continuous Access replication tunnels Replicated EVA Vdisks (copy sets) are grouped in DR groups. When a DR group is created, a relationship is es
25 For best performance of remote replication, the average workload (reads and writes) should be equal on both controllers and on both fabrics and IS
26 Sizing the SAN ISL The location of the remote site often determines the ISL technologies that meet your distance and performance requirements. The
27 Protection Group 2 Microsoft SQL Server 2008 plus IIS 3 Protected Virtual Machines: SQL Server 2008, IIS and a Windows 2008 application clie
28 In our environment, MS Exchange 2007 and Linux virtual machines run at Site A and replicate to Site B. MS SQL Server 2008, IIS and the file server
29 In previous testing we established that VMFS and Raw Device Mapping perform equally well when subjected to an Exchange 2007 workload. In this conf
3 HP StorageWorks Continuous Access EVA Continuous Access EVA is a feature of the Enterprise Virtual Array (EVA) that allows data replication betwee
30 Figure 18. Full copy using a single DR group Figure 19. Full copy using two DR Groups As shown in Table 4, spreading the application data
31 Failover considerations Failover occurs when a Site Recovery Manager recovery plan is executed, and the Site Recovery Manager recovery plan is con
32 The impact of test failovers that include application recovery can be mitigated with adequate disk array sizing. These additional IOPS must be fac
33 Failback considerations Failback is the process of setting the replication environment back to its original state at the protected site prior to f
34 Best practices and caveats This section includes recommendations for customers who plan to deploy VMware vCenter Site Recovery Manager, using HP S
35 Storage administration EVA disk group layout: To optimize performance and provide fault tolerance, place application databases files, transacti
36 Conclusion The test results and related information contained in this paper clearly illustrate how to properly plan for and successfully deploy a
37 For more information HP solutions and white papers HP StorageWorks SAN Design Reference Guide http://h20000.www2.hp.com/bc/docs/support/SupportM
Technology for better business outcomes © Copyright 2009 Hewlett-Packard Development Company, L.P. The information contained herein is subject to cha
4 Solution Scope Building an effective disaster tolerant solution can often be a very complex and time consuming task. Furthermore most disaster tol
5 Figure 1. Configuration diagram In this configuration, the VMware ESX servers can be clustered so that failures of a single ESX host only requir
6 Software and hardware configuration The following software and hardware components are used in this case study. Software VMware ESX server 3.5
7 Figure 2. Zoning example SAN switch settings for Continuous Access We used the following fabric Brocade Switch settings, which are based on rule
8 To determine the number of spindles required for performance, the following simple formula can be used: Number of drives needed = (Total IOPs * RAI
9 For example, a disk group configuration using 300 GB drives, single sparing, and Vraid 5 for a total estimated VM disk space of 8 TB breaks down as
Comments to this Manuals