Rick Andersen

Hitachi Storage Cluster with Windows 2012 R2

Discussion created by Rick Andersen on Jan 23, 2014
Latest reply on Jun 11, 2014 by Herman Rutten

 

Hitachi Data Systems, working closely with Microsoft®, optimizes storage solutions deployed using for Microsoft Windows Server® 2012 R2 using Microsoft Hyper-V®. Integrating solutions using storage array replication with Windows Server 2012 R2 and Hyper-V provide high availability for disaster recovery scenarios.

Disaster recovery solutions consist of remote LUN replication between two sites. The automated failover of the guest virtual machines resources to the secondary site occur if the main site goes down or is otherwise unavailable.

Hitachi Storage Cluster for Microsoft® Hyper-V® handles data replication and control, and the storage array controllers. This has little impact on the applications running in the guest machine and is fully automated and controlled by the cluster. Consistency groups and time-stamped writes ensure database integrity.

Virtual guest machines run as clusters resources within the Hyper-V cluster. If a node hosting a virtual guest machine within the cluster fails, the virtual guest machine automatically failovers to an available node. The virtual machines can be moved quickly between cluster nodes to allow for planned and unplanned outages. Storage Cluster automatically brings the replicated LUNS and the virtual machine online by the cluster.

Storage Cluster provides a disaster recovery solution for Microsoft Windows 2012 R2. It “cluster-enables” disk resources that have been replicated to a distant site using Hitachi TrueCopy. Since there is a recovery-suitable disk image available at a remote location, the cluster can fail over to the recovery facility and pick up processing using that I/O-consistent image of production disks.

Storage Cluster “stretches” a cluster across distance. Production disk resources are replicated to the recovery site. Hitachi Storage Cluster ensures that the recovery nodes have appropriate access to the disk resources of the cluster to resume operation with a standard cluster failover. This enables real-time cluster recovery of critical applications across widely separated geographic locations.

This solution uses the following:

§  Two Hitachi Unified Storage VMs

§  Two Hitachi Compute Blade 500 blades with Microsoft Windows Server 2012 R2 in a Hyper-V failover cluster.

A use case tests the live migration of SQL virtual machines between the two stretched nodes in the failover cluster.

To enable Hitachi Storage Cluster with Windows 2012 R2 in a Hyper-V failover cluster, perform the following steps:

1.     Configure Hitachi Storage Cluster. This software consists of a generic resource script which controls the replication of LUNs on a virtual machine in the cluster.

§  All LUNs used by the replicated virtual machines must reside on VHDx volumes within the cluster.

2.     Store the virtual machine with its VHDx and configuration files on a single LUN.

§  Add these with either drive letters or mount points. Recommended practice is to use mount points. This keeps the number of drive letters from being a limitation.

§  Mount the VHDx disk to folders on drive C: of the respective Hyper-V cluster nodes. .

3.     Locate any additional disks for the virtual machine so that the data and log files are on their own LUN and configured with mount points similar to the virtual machine files.

§  Instead of adding the additional disks to the virtual machine as pass-through disks, store the data and log files as VHDx files on separate LUNs.

§  Add these disks to the virtual machine using mount points.

4.     Configure the replication environment.

5.     Make the virtual machines that are part of a replication group highly available within the cluster.

The first test case used a clustered instance of SQL Server to validate the success of doing the following across stretched nodes in the failover cluster:

§  Live migration

§  Quick migration

The second test case used a SQL Server virtual machine hosted in a Hyper-V failover cluster to validate the success of doing a Live Migration and a Quick Migration across stretched nodes in the failover cluster. This SQL Server virtual machine was setup as a highly available virtual machine in the Hyper-V failover cluster.

Table 1. Test Cases

Test case

Pass/fail criteria

Result

SQL Server Clustered Instance High  Availability test using Hitachi Storage Cluster

Successful live migration and quick   migration

Pass.

Standalone SQL Server Virtual Machine   High Availability test using Hitachi Storage Cluster

Successful live migration and quick   migration

Pass

Hitachi Storage Cluster provides the following benefits:

§  Extends Microsoft Cluster Service with a Hitachi TrueCopy resource or a Hitachi Universal Replicator resource to monitor and manage the replicated data (groups of disks).

§  Coordinates replication activity (replication takeover) with cluster operations (online, move, etc.).

§  No additional cluster commands to learn or manage.

§  Straightforward configuration.

§  Uses Hitachi TrueCopy or Hitachi Universal Replication for robust off-host disk replication to the remote facility.

This document provides the following:

§  A proof point of the basic functionality of this solution

§  High level technical reference for considering this solution

§  High level reference of the use case implementation

This document does not cover the following:

§  Performance measurement

§  Sizing information

§  Best practice

§  Implementation details

For implementation details, contact your Hitachi Data Systems representative.

NoteTesting of this configuration was in a lab environment. Many things affect production environments beyond prediction or duplication in a lab environment. Follow the recommended practice of conducting proof-of-concept testing for acceptable results in a non-production, isolated test environment that otherwise matches your production environment before your production implementation of this solution.

Tested Components

This describes the hardware and software components used to test Hitachi Storage Cluster for Microsoft Hyper-V.

Table 2 Hardware Components

Hardware

Description

Version

Quantity

Hitachi Unified Storage VM

§  Dual controllers

§  16 × 8 Gb/sec Fibre Channel ports

§  64 GB cache memory

§  72 × 600 GB 10k RPM SAS disks, 2.5 inch SFF

73-02-00-00/01

2

Hitachi Compute Blade 500 chassis

§  2-blade chassis

§  2 Brocade 5460 Fibre Channel switch   modules, each with 6 × 8 Gb/sec uplink ports

§  2 Brocade VDX 6746 Ethernet switch   modules, each with 8 × 10 Gb/sec uplink ports

§  2 management modules

§  6 cooling fan modules

§  4 power supply modules

SVP: A0145-H-7304

5460: FOS   6.3.2d1

VDX6746:   NOS 2.0.1_kat4

1

520BH1 server blade

§  Half blade

§    2 × 8-core Intel Xeon E5-2680   processor, 2.70 GHz

§  96 GB RAM

BMC/EFI: 01-59

2

Table 3. Software Components

Software

Version

Hitachi Storage Navigator

Microcode Dependent

Hitachi Raid Manager

CCI-01-30-03

Windows Server 2012

R2

SQL Server

2012

High Level Test Infrastructure

Figure 1 is the high-level test architecture employed to test Hitachi Storage Cluster for Microsoft Hyper-V using Microsoft Windows 2012 R2 in a Hyper-V failover clustering. This deployed a SQL server virtual machine to test the live migration and quick migration between a geographically dispersed fail over cluster.

  HSC-Hyper-V with R2 Comm.jpg                     

Figure 1

Test Result

Table 4 describes the test result and success criteria for this testing

Table 4. Test Results

Test case

Description

Result

Clustered SQL Server Failover

Successful Live Migration and Quick Migration   of clustered SQL virtual machines

Passed.

Standalone SQL Server Virtual Machine   Failover

Successful Live Migration and Quick   Migration of standalone SQL Virtual Machine

Passed

Outcomes