ucp srm whitepaper

White Paper

By UCP Engineering Technology Team

September 2012

i

Table of Contents

Introduction ................................................................................................................................................................................................. 2

Tested Solution Components ..................................................................................................................................................................... 3

Hardware Components .................................................................................................................................................................... 3

Hitachi Virtual Storage Platform ............................................................................................................................................ 4

Hitachi Compute Blade 500 .................................................................................................................................................. 4

Brocade Ethernet Network .................................................................................................................................................... 5

Brocade Fibre Channel Fabric .............................................................................................................................................. 5

Software Components ..................................................................................................................................................................... 6

Hitachi Unified Compute Platform for VMware vSphere integrated solution ......................................................................... 7

Hitachi TrueCopy Heterogeneous Remote Replication Bundle............................................................................................. 7

Hitachi Universal Replicator .................................................................................................................................................. 7

Hitachi ShadowImage Heterogeneous Replication ............................................................................................................... 8

VMware vSphere 5.1 ............................................................................................................................................................ 8

VMware vCenter Site Recovery Manager 5.1 ....................................................................................................................... 8

Solution Implementation ............................................................................................................................................................................. 9

Configure the Storage Area Network ............................................................................................................................................... 9

Configure the Storage Replication Link ......................................................................................................................................... 12

Hitachi TrueCopy Remote Replication Bundle .................................................................................................................... 13

Hitachi Universal Replicator ................................................................................................................................................ 13

Creating Journal Groups for Hitachi Universal Replicator ................................................................................................... 14

Configure Storage.......................................................................................................................................................................... 14

Create and Attach Datastore to Host via UCP .................................................................................................................... 15

Create LDEVs within the Management Pool (Journal Vol, & Command Device) ................................................................ 16

Add Volumes to Management Node HSDs ......................................................................................................................... 16

Create a Command Device ................................................................................................................................................. 16

UCP Pro for VMware vSphere SRM environment ......................................................................................................................... 17

Configure Command Control Interface (CCI) for Replication ......................................................................................................... 19

Install Command Control Interface ...................................................................................................................................... 19

Configure Command Devices ............................................................................................................................................. 20

Configure Remote Replication Using Command Control Interface ..................................................................................... 20

(Optional) Configure Local Replication Using Command Control Interface ........................................................................ 22

Implement Storage Replication for Site Recovery Manager .......................................................................................................... 23

Enabling replication for TrueCopy Synchronous ................................................................................................................. 24

Enabling replication for Universal Replicator....................................................................................................................... 24

(Optional) Enabling replication for ShadowImage ............................................................................................................... 24

Configure Storage Replication Adapter with VMware Site Recovery Manager .............................................................................. 24

Other Implementation Actions ....................................................................................................................................................... 26

Configuration Files ................................................................................................................................................................................... 27

Hitachi TrueCopy Remote Replication Bundle Files ...................................................................................................................... 27

Hitachi Universal Replicator Files .................................................................................................................................................. 28

Hitachi ShadowImage Heterogeneous Replication Files ............................................................................................................... 29

For More Information ................................................................................................................................................................................ 31

2

Introduction

When deploying a virtualized data center with for VMware vSphere, there are several benefits gained in IT infrastructures:

Pre-configured and pre-tested Converged Infrastructure solution

Integrated infrastructure management capabilities directly into vSphere client

Reduce operating expenses through automated UCP workflows to allow a data center to run more efficiently

Protect critical applications running on a converged infrastructure with features such as VMware High Availability, which reduces planned and unplanned downtime

Expanding data centers across multiple locations provides an opportunity to increase these layers of protection beyond a single data center.

VMware vCenter Site Recovery Manager is a business continuity and disaster recovery solution that integrates VMware vSphere infrastructures with data replication on storage systems to protect large sites and business critical applications. It does the following:

Provides centralized management of disaster recovery plans

Enables non-disruptive testing

Automates site recovery and migration processes

Using an automated and tested recovery plan, a production data center can do a disaster recovery failover in an organized and proven manner.

Remote data replication is a key function in building out stable and reliable disaster recovery environments. Replicating data to a remote secondary site represents the most effective insurance policy against a catastrophic failure. Although you can perform data replication at the server level, you can perform data replication more effectively within the storage infrastructure.

Hitachi Virtual Storage Platform is an integral piece in building out a robust business continuity and disaster recovery solution. VMware vCenter Site Recovery Manager integrates tightly with Hitachi Virtual Storage Platform using the Hitachi Storage Replication Adapter. The advanced functions found in the Virtual Storage Platform fulfill the requirements of a virtual infrastructure and provide reliable protection by managing data replication across data centers.

This paper's intended use is by IT administrators charged with the storage, deployment, or administration of platform. It assumes familiarity with storage area network (SAN)-based storage systems, VMware vSphere, Hitachi data replication technologies, and common IT storage practices.

3

Tested Solution Components

This white paper describes how to implement a disaster recovery solution for a VMware virtualized infrastructure using Hitachi hardware and software. Specifically, this solution uses which contain the following:

Hardware Components

The following tables describe the hardware needed to deploy this solution. The UCP Pro solution for each site includes all of

the components listed.

Hardware Quantity Configuration Role

Hitachi Compute Blade 500 chassis

1 8 x 520H B1 blades, Stateless ESXi Compute Nodes

Hitachi CR210 1u servers

2 Boot-from-SAN, Configured as a VMWare HA Cluster

UCP Pro management services/VMs

Brocade 5460 2 In-chassis Fibre Channel switches Edge switch (Core-Edge SAN model)

Brocade 6510 2 Fibre channel switches

Three vFabs configured for; o Compute resources o Management resources o Site-to-site replication

Core Switch (Core-Edge SAN model)

Brocade 6746 2 In-chassis Ethernet switches Access layer

Brocade 6720 2 Ethernet switches Aggregation layer

Table 1. Base UCP Pro Compute Rack

Hardware Quantity Configuration Role

Hitachi Virtual Storage Platform (VSP)

1 16 x 8Gb/sec Fibre Channel ports

(4) Management Host Ports

(8) Compute Host Ports

(4) Replication Ports

Primary and remote storage

Table 2. Base UCP Pro Storage Rack

o Vcenter Server 5.1 o SQL 2012 o o UCP Director o Hitachi Command Suite (HCSM & HDVM) o SRM 5.1

The software for storage replication using the following: o Hitachi Command Control Interface o Hitachi TrueCopy Remote Replication bundle o Hitachi Universal Replicator

4

The figure below illustrates a high- Detailed cabling designs within a base compute

UCP Hardware Assembly & Configuration Manual {official doc number here}), while intra-site connectivity and configuration will be detailed later in this paper.

= 8Gb/sec FC

= 16Gb/sec FC ISL

= 10 Gb/sec IP

LAN/WAN

Figure 1 UCP Site-to-site

Hitachi Virtual Storage Platform

Virtual Storage Platform is the first 3D scaling storage platform designed for all data types. Its storage architecture flexibly adapts for performance, capacity, and multi-vendor storage. Combined with the unique Hitachi Command Suite management platform, it transforms the data center.

Scale Up Meet increasing demands by dynamically adding processors, connectivity, and capacity in a single unit. Provide the highest performance for both open and mainframe environments.

Scale Out Meet multiple demands by dynamically combining multiple units into a single logical system with shared resources. Support increased demand in virtualized server environments. Ensure safe multi-tenancy and quality of service through partitioning of cache and ports.

Scale Deep Extend storage value by virtualizing new and existing external storage systems dynamically. Extend the advanced functions of Hitachi Virtual Storage Platform to multivendor storage. Offload less demanding data to external tiers to save costs and to optimize the availability of tier one resources.

For more information, see Hitachi Virtual Storage Platform on the Hitachi Data Systems website.

Hitachi Compute Blade 500

Hitachi Compute Blade 500, an enterprise-class blade server platform, is a key component of the UCP Pro converged infrastructure. It features the following:

High VM-to-blade density ratio making it an ideal virtualization platform

Latest Intel Xeon processors for state-of-the-art performance

Configuration flexibility

Eco-friendly power-saving capabilities

5

For more information, see Hitachi Compute Blade Family on the Hitachi Data Systems website.

Brocade Ethernet Network

Brocade and Hitachi Data Systems partnership to deliver key networking and data center solutions.

The Ethernet infrastructure that is implemented in the UCP Pro converged infrastructure is designed to deliver high I/O throughput of a 10GbE network while providing full redundancy for IP traffic.

Brocade VDX 6720 is a high-performance, ultra-low latency wire-speed 10 Gigabit Ethernet (GbE) fixed configuration switch. The Brocade VDX 6720 is an ideal platform for a variety of Top-of-Rack (ToR) fabric deployments. Within the UCP Pro base compute rack,

Brocade VDX 6746 is an OEM, ultra-low latency wire-speed 10 Gigabit Ethernet (GbE) in-chassis switch. For each CB 500 Compute Chassisthe top of the rack of the UCP Pro base compute rack.

The figure below shows the connections between;

The blades in the chassis and the 6746 in-chassis switches,

The 6746 in-

The Link Aggregation Groups (LAG)

The uplink LAG to the customer network

Figure 2 UCP Ethernet connectivity

Brocade Fibre Channel Fabric

The Fibre channel infrastructure that is implemented in the UCP Pro converged infrastructure follows Hitachi and Brocade best practices, and offers redundancy within the fabric, and multiple paths between host and volume for maximum throughput and availability.

The Brocade 6510 fibre channel switch meets the demands of hyper-scale, private cloud storage environments by delivering market-leading 16 Gbps Fibre Channel technology and capabilities that support highly virtualized environments.

The Brocade 5460 in-chassis fibre channel switch is an OEM developed switch based on the popular 5000-series FC switch line.

6

The figure below shows the connections between;

The blades in the chassis and the 5460 in-chassis switches,

The 5460 in-chassis switches and the 6510

The Storage Array Ports to the 651

Figure 3 Fibre channel connectivity

Software Components

The table below lists the software needed to deploy this solution.

Software Version

Hitachi UCP Pro (major components listed below)

VMware vCenter 5.1 build 799731

VMware vSphere ESXi 51 build 799733

UCP Director 2.0.0 build 1711

Hitachi Device Manager 7.3.1-02

Hitachi Compute Server Manager 7.3.0-02

7

VMware Site Recovery Manager 5.1 build 820150

Hitachi Storage Replication Adapter 02.01.3

Hitachi Command Control Interface 01-28-03/05

Note: Ver. 01-24/03/13 or later is required

Hitachi Storage Navigator Microcode dependent

Hitachi TrueCopy Remote Replication bundle Licensed feature available with Hitachi Virtual Storage Platform

Hitachi Universal Replicator Licensed feature available with Hitachi Virtual Storage Platform

Hitachi ShadowImage Heterogeneous Replication Licensed feature available with Hitachi Virtual Storage Platform

Table 2. Solution Software

Hitachi Unified Compute Platform for VMware vSphere integrated solution

Hitachi Unified Compute Platform Pro for VMware vSphere (UCP Pro for VMware vSphere) centralizes and automates the management of servers, storage and networking components as business resource pools. Its comprehensive and simplified management significantly reduces time to value and operational costs for even the most demanding applications. UCP Pro for

resource allocation. It also supports simple automation and orchestration for streamlined infrastructure management.

UCP Pro for VMware vSphere is an exceptionally flexible, converged solution that simplifies deployment and management of data centers and private cloud deployments through and software with comprehensive management.

UCP Pro for VMware vSphere is designed for maximum performance and utilization leveraging Hitachi best-in-class hardware

reducing deployment time.

Hitachi TrueCopy Heterogeneous Remote Replication Bundle

For synchronous replication up to 190 miles (300 km) with WAN optimized controllers, Hitachi TrueCopy Heterogeneous Remote Replication bundle provides a no-data-loss, rapid restart solution. Real-time copies are the same as the originals. This reduces recovery time to minutes.

Synchronous replication provides very fast recovery time (low RTO) and good data currency (low RPO) between Hitachi Data Systems storage systems.

For more information, see TrueCopy Synchronous Remote Replication Software on the Hitachi Data Systems website.

Hitachi Universal Replicator

Hitachi Universal Replicator for the Hitachi Virtual Storage Platform is an advanced technology for asynchronously replicating data hosted on Virtual Storage Platform models.

Asynchronous replication maintains a remote copy of data without having a performance impact on the primary data. In asynchronous replication, the system considers an I/O update operation complete when confirming the update at the primary site.

The Hitachi Virtual Storage Platform manages the process of replicating the changes to the secondary site.

Hitachi Universal Replicator software uses disk-based journaling and an optimized replication engine to reduce resource consumption and costs while increasing performance and operational resilience. The strengths of Hitachi Universal Replicator are two key technical innovations: performance optimized, disk based journaling and a pull-style replication engine.

The following are the key characteristics of Universal Replicator:

Replication across any distance without significant negative effect on host performance Hitachi Universal Replicator has been successfully used in replication configurations that span thousands of miles.

8

No acknowledgement dependencies from secondary site Hitachi Universal Replicator replicates to a remote site without the performance impact of waiting to acknowledge each individual record. Instead, Hitachi Universal Replicator manages the remote relationship at a controller level. During a disruption of communication to the remote unit or while exceeding the capability of the replication circuit, Hitachi Universal Replicator retains replicated write data in local journals, then updates the write data when the condition is corrected.

There is a potential for some data lag between remote and primary sites, particularly at longer distances. Manage the recovery point objective with the configuration of the data communication lines. When I/O activity at the primary site exceeds the capacity of the communication channel, the data is staged and moved to the secondary site in the same order as it was written at the primary site.

Hitachi ShadowImage Heterogeneous Replication

Hitachi ShadowImage Heterogeneous Replication is a storage-based hardware solution that creates RAID-protected duplicate volumes within Hitachi Virtual Storage Platform. ShadowImage Heterogeneous Replication primary volumes (PVOL) contain the original data. Up to nine secondary volumes (S-VOL) can be created as copies.

On Hitachi Virtual Storage Platform, use ShadowImage Heterogeneous

Replication to implement clones, a full copy of the primary data. Each clone is available for use by secondary applications. The unique value of working with a clone is that any operation on a clone has no effect on the primary data.

Detailed information on using Shadow Image Heterogeneous Replication is in Hitachi Command Control Interface User and Reference Guide (MK-90RD7010).

VMware vSphere 5.1

VMware vSphere 5.1 is a virtualization platform that provides a data center infrastructure. It features Distributed Resource Scheduler (DRS), high availability, and fault tolerance.

Use of the round robin multipathing policy in vSphere 5.1 distributes the load across multiple host bus adapters (HBAs) and multiple storage ports. Use of DRS with Hitachi Dynamic Provisioning automatically distributes loads on the ESX host and across the back end of the storage system.

VMware vSphere 5.1 has the following components:

ESXi 5.1 This is a hypervisor that loads directly on a physical server. It partitions one physical machine into many virtual machines that share hardware resources.

Stateless ESXi where the hypervisor is loaded directly into memory, and is not booted from disk.

vCenter Server This allows management of the vSphere environment through a single user interface. With vCenter, there are features available such as vMotion, Storage vMotion, Distributed Resource Scheduler, high availability, and fault tolerance.

For more information, see the VMware vSphere website.

VMware vCenter Site Recovery Manager 5.1

VMware vCenter Site Recovery Manager 5.1 is a disaster recovery solution that helps to reduce planned and unplanned downtime of a vSphere infrastructure. It enables automated site recovery and migration processes. This can leverage the built-in vSphere Replication for hypervisor-based replication to cover a wide range of required recovery time and date currency.

This reference architecture focuses on using Site Recovery Manager with storage-based replication technologies such as Hitachi TrueCopy Heterogeneous Remote Replication bundle and Hitachi Universal Replicator. This use provides centralized management of recovery plans. Tight integration between storage arrays, VMware vCenter, VMware vCenter Site Recovery Manager, and the Hitachi Storage Replication Adapter ensure a coordinated recovery for large, business critical environments.

For more information, see VMware vCenter Site Recovery Manager on the VMware website at http://www.vmware.com/products/site-recovery-manager/resource.html#docs .
http://www.vmware.com/products/site-recovery-manager/resource.html#docs

9

Solution Implementation

s with Compute Blade 500 chassis that are fully populated with all 8 blades, the following section will focus on demonstrating replication for a single host, however the steps and concepts can be expanded to include all blades within the site.

Follow these steps to deploy this solution:

1. Configure the Storage Area Network for virtual fabric and replication links 2. Configure the Storage Replication Link 3. Configure Storage 4. Configure vSphere environment 5. Configure Command Control Interface (CCI) for Replication 6. Implement Storage Replication for Site Recovery Manager

Configure the Storage Area Network

The 6510 Fibre Channel switches are configured with virtual fabrics to segregate general compute host traffic from management host traffic. In addition to the UCP Pro virtual fabrics, an additional virtual fabric is needed for the replication links in this solution. The following diagram displays the virtual fabrics within the 6510 switches.

Brocade6510

VSP

Vfab 128

Vfab 1

Vfab 2

Vfab 128

Vfab 1

Vfab 2

CL1-A, CL2-A

CL3-A, CL4-A

CL7-A, CL8-A

CL5-A, CL6-A

CL1-B, CL2-B,CL5-B, CL6-B


Brocade6510

Vfab 128

Vfab 1

Vfab 2

Vfab 128

Vfab 1

Vfab 2

VSP

CL1-A, CL2-A

CL3-A, CL4-A

CL7-A, CL8-A



= 8Gb connection

= 16Gb x 2 Channel ISL

CL5-A, CL6-A

Figure 4 - Virtual fabric connectivity

10

Each server blade at the local and remote site uses dual-port Fibre Channel mezzanine cards. They connect internally to the internal Fibre Channel switch modules located in the Hitachi Compute Blade 500 chassis as depicted earlier in figure 3.

Four inter-switch links from each internal Fibre Channel switch modules connect to a Brocade 6510. In turn, each Brocade 6510 switch connects to eight target ports on the Hitachi Virtual Storage Platform.

The Hitachi Virtual Storage Platform supports active-active multipath connectivity. UCP Best practices ensures that at least four unique paths exist from the ESXi host to the storage system to maximize availability. The multipathing policy was set to round robin in ESXi 5.1.

The tables below show the zoning configuration for the protected site and the recovery site.

Host Host HBA/Port Fabric Zone Name Storage Port

Protected Site Compute1

HBA Port 0 6510 #1

vFab #128

Port0_Compute1_CL1B_VSP_10_20_90_66 CL1B

Protected Site Compute 1

HBA Port 0 6510 #1

vFab #128

Port0_ Compute1_CL2B_VSP_10_20_90_66 CL2B


HBA Port 1 6510 #2

vFab #128



HBA Port 1 6510 #2

vFab #128


Protected Site Mgmt1

HBA Port 0 6510 #1

vFab #1

Port0_Mgmt1_ CL1A_VSP_10_20_90_66 CL1A


HBA Port 0 6510 #1

vFab #1



HBA Port 1 6510 #2

vFab #1



HBA Port 1 6510 #2

vFab #1



HBA Port 0 6510 #1

vFab #1



HBA Port 0 6510 #1

vFab #1



HBA Port 1 6510 #2

vFab #1



HBA Port 1 6510 #2

vFab #1


Table 1 Site A zoning, Management virtual fabric (1) & Compute virtual fabric (128)

11

Host Host HBA/Port Fabric Zone Name Storage Port

Recovery Site Compute1

HBA Port 0 6510 #1

vFab #128

Port0_Compute1_CL1B_VSP_10_20_90_67 CL1B

Recovery Site Compute 1

HBA Port 0 6510 #1

vFab #128



HBA Port 1 6510 #2

vFab #128



HBA Port 1 6510 #2

vFab #128


Recovery Site Mgmt1

HBA Port 0 6510 #1

vFab #1


Recovery Site Mgmt1

HBA Port 0 6510 #1

vFab #1


Recovery Site Mgmt1

HBA Port 1 6510 #2

vFab #1


Recovery Site Mgmt1

HBA Port 1 6510 #2

vFab #1


Recovery Site Mgmt2

HBA Port 0 6510 #1

vFab #1


Recovery Site Mgmt2

HBA Port 0 6510 #1

vFab #1


Recovery Site Mgmt2

HBA Port 1 6510 #2

vFab #1


Recovery Site Mgmt2

HBA Port 1 6510 #2

vFab #1


Table 2 Site B zoning, Management virtual fabric (1) & Compute virtual fabric (128)

In addition, replication requires dedicated Fibre Channel connections between storage systems at the local and the remote site. For this solution, each Hitachi Virtual Storage Platform uses a total of two initiator ports and two RCU target ports.

The following table and diagram shows the storage replication paths and zoning between storage systems on each site.

12

Storage System Storage Port Fabric Zone Name Storage System Storage Port

Protected Site - VSP1

5A - Initiator 6510 #1 (both sites)

vFab #2

VSP1_5A_VSP2_5A Recovery Site - VSP2

5A - RCU Target


6A - RCU Target

6510 #1 (both sites)

vFab #2


6A - Initiator


7A - RCU Target

6510 #2 (both sites)

vFab #2


7A - Initiator


8A - Initiator 6510 #2 (both sites)

vFab #2


8A - RCU Target

Table 3 Virtual Fabric zoning for replication ports

VSP Site BVSP Site A

5AInitiator

6ARCU

Target

7ARCU

Target

8AInitiator

5ARCU

Target

6AInitiator

7AInitiator

8ARCU

Target

Figure 5 - Initiator/Target configuration

Configure the Storage Replication Link

Site Recovery Manager supports two types of replication technologies available with Hitachi Virtual Storage Platform. Choose the storage replication technology based on the recovery point object (RPO) requirement, as shown in the table below.

13

Table 4 RPO Requirements

Requirement Replication Type Storage Replication Technology

Low RPO Synchronous remote replication Hitachi TrueCopy Heterogeneous Remote Replication Bundle

Flexible RPO Asynchronous remote replication Hitachi Universal Replicator

The following describes how to configure the storage replication link between the local and remote storage systems for the respective replication technologies.

Hitachi TrueCopy Heterogeneous Remote Replication Bundle


Creating Journal Groups for Hitachi Universal Replicator

Hitachi TrueCopy Remote Replication Bundle

This procedure assumes that zoning storage replication paths within your Fibre Channel fabric is complete. To create storage replication links for Hitachi TrueCopy on Virtual Storage Platform using Hitachi Storage Navigator software, follow these steps:

1 From the Actions menu, point to Remote Copy, then point to TrueCopy, and then click RCU Operation. 2 Modify settings on the storage ports.

1 Click the Pen button in the top right area of the pane to enter Modify mode. 2 Click the Port option. 3 Right-click the storage ports to change the port attribute to Initiator or RCU Target, as appropriate. This solution uses

ports for Initiator and RCU Target as depicted above in figure 5. 4 Click the MCU&RCU option. 5 Click CU Free from the navigation tree. TrueCopy software selects the first available control unit to be the RCU. 6 Right-click the blank space on the right pane. A shortcut menu opens.

3 Add the RCU (Fibre) connections. 1 Click RCU Operation and then click Add RCU (Fibre) from the shortcut menu. The Add RCU (Fibre) dialog box

opens. 2 Provide the following information in the Add RCU (Fibre) dialog box:

1 Type the serial number for the remote storage system in S/N. 2 Click the LDKC of the remote storage system from the LDKC list. 3 Click 6 (VSP) from the Controller ID list. 4 Select the Default check box for the Path Group ID. 5 Click the local storage system port and its corresponding pair port in the MCU-RCU Path area. This solution uses

ports for Initiator and RCU Target as depicted above in figure 5. 3 Set options.

1 Click Option. 2 Type values in Minimum Paths, RIO MIH Time (sec.) and Round Trip Time (ms). 3 Click Set.

4 Accept the changes and complete the changes. 1 Click Apply. A confirmation dialog box opens. 2 Click OK to confirm applying the changes. Another confirmation dialog box opens. 3 Click OK.

Repeat these steps on the remote storage system to add an RCU (Fibre) connection for reverse replication. This is necessary for VMware Site Recovery Manager to perform failback operations.


This procedure assumes that zoning storage replication paths within your Fibre Channel fabric is complete. To create storage replication links for Hitachi Universal Replicator on Virtual Storage Platform using Hitachi Storage Navigator software, follow these steps:

1 From the Actions menu, point to Remote Copy, then point to Universal Replicator, and then click DKC Operation. 2 Modify settings on the storage ports.

1 Click the Pen button in the top right area of the pane to enter Modify mode. 2 Click the Port option. 3 Right-click the storage ports to change the port attribute to Initiator or RCU Target, as appropriate. This solution uses

ports for Initiator and RCU Target as depicted above in figure 4. 4 Click the DKC option. The LDKC navigation tree displays in the left pane. 5 Click LDKC#00 on the navigation tree.

14

6 Right-click the blank space on the left pane. A menu displays. 3 Add the DKC connections.

1 Click DKC Operation and then click Add DKC from the list. The Add DKC dialog box opens. 2 Provide the following information in the Add DKC dialog box:

1 Type the serial number for the remote storage system in S/N. 2 Click the LDKC of the remote storage system from the LDKC list. 3 Click 6 (VSP) from the Controller ID list. 4 Click the local storage system ports and its corresponding pair-port in the M-R path area.

3 Set options. 1 Click Option. 2 Type values in Minimum Paths and RIO MIH Times (sec). 3 Click Set.

4 Accept the changes and complete the changes. 1 Click Apply. A confirmation dialog box opens. 2 Click OK to confirm applying the changes. Another confirmation dialog box opens. 3 Click OK.

Repeat these steps on the remote storage system to add a DKC connection for reverse replication. This is necessary for VMware Site Recovery Manager to perform failback operations.

Creating Journal Groups for Hitachi Universal Replicator

To create Hitachi Universal Replicator journal groups using Hitachi Storage Navigator, follow these steps on the local storage system:

1 From the Actions menu, point to Remote Copy, then point to Universal Replicator, and then click Journal Operation. 2 Modify settings for the journals.

1 Click the Pen button in the top right area of the pane to enter Modify mode. 2 In the Journals tree, click to expand Free. 3 Select (highlight) a free journal group from Free. Details of the selected free journal group display in the right area. 4 Right-click the highlighted group in the right area, and then click Edit JNL VOLs from the shortcut menu. The Edit JNL

Volumes dialog box opens. 5 Populate the journal volumes.

1 Click the CU option. 2 From the menu on the right, click a CU. The Free Volumes pane populates.

6 Populate the journal volumes. 1 Select one or more volumes in the Free Volumes area. The Add button becomes available. 2 Click Add. The JNL Volumes area populates.

3 Save the changes. 1 Click Set and then click Apply. A confirmation dialog box opens. 2 Click Yes. A final confirmation opens. 3 Click OK.

Repeat these steps on the remote storage system to add a journal group.

Configure Storage

This is how to configure your storage for this solution using Hitachi Dynamic Provisioning on Hitachi Virtual Storage Platform. The storage configuration for this solution includes the following:

Over provisioning

Wide striping

On-line expansion of dynamic provisioning pools

This solution uses a dynamic provisioning pool comprised of a single RAID group with eight 600 GB 10k RPM SAS drives in a RAID-6 (6D+2P) configuration for each storage system. To increase performance and capacity needs of your virtual environment, add more RAID groups to the dynamic provisioning pool.

Using a RAID-6 configuration lowers the risk of data loss or pool failure, which is a primary concern for virtual machines protected by VMware Site Recovery Manager. Refer to Hitachi VSP Architecture guide for further details on Pool configuration/benefits at: http://www.hds.com/assets/pdf/hitachi-architecture-guide-virtual-storage-platform.pdf .

The figure below shows the configuration for the dynamic provisioning pool on each storage system. For simplicity sake of this paper, LDEV IDs are identical between sites.
http://www.hds.com/assets/pdf/hitachi-architecture-guide-virtual-storage-platform.pdf

15

Management Pool

Datastore1

Datastore2

RAID 6 (6+2) PG 1-1

Compute Pool

CMD

RAID 6 (6+2) PG 1-3

RAID 6 (6+2) PG 1-5

Journal Vols*

Datastore

*Journal Volumes are only required w/ Hitachi Universal Replicator

Jnl

Figure 6 Pool and volume layout

The UCP Pro solution at each site will have the Management Pool and volumes pre-configured at the factory will contain 2 shared Datastores where the UCP Management virtual machines will be stored.

The compute pool will be configured to customer specifications onsite during the UCP system installation and configuration. Upon completion of the UCP Pro system, the entire Server/Storage/Network stack will be in an operational state. For purposes of this paper, this solution provisions 3 additional LDEVs to the VMware infrastructure at both sites for:

Datastore volume for compute host

Journal volume (only if HUR replication is used)

Command device (raw device mapping LUN to SRM/CCI server)

The creation and presentation of the LDEVs in Figure 3 on the storage system involves the following steps:

1 Create and Attach Datastore to Host via UCP 2 Manually Create and configure LDEVs in Compute pool

Journal Volume

Command Device 3 Attach Existing Volume to via UCP

The following describe in detail how to configure your storage for this solution.

Create and Attach Datastore to Host via UCP

This procedure creates the proper zoning within the Fibre Channel infrastructure, the volume, the host storage domain (HSD) and the HSD mappings to the ESXi host HBA's World Wide Names (WWN). Host groups are then associated with LDEVs to ensure access is isolated to the specified hosts.

To create and attach a datastore to a host using Hitachi UCP Pro, follow these steps:

16

1 From within vSphere client, right-click on the server that you want to add a volume to, and then click on the Configure Host Storage link.

2 Select the Create New Volume option. 3 Click on the Next button. 4 On the Create New Volume step:

In the Name field, type a name to assign to the volume.

In the Disk Size field, type the total amount of space to use to create the volume in GB.

To format the volume as a datastore, select the Format as Datastore option. If this option is not selected, the

created volume will need to be formatted later before it can be used.

To manually select the array ports to use, select the Manually SelectArray Ports option, and then select the

appropriate ports for each fabric.

In the Pools table, select the pool to create the data store from. 5 Click on the Finish button.

6 If this is the first and only volume created within the array so far, the volume number (LDEV) should be 00:00:00. Verify the LDEV ID, and if it is not 00:00:00, please record the actual value to be used later in the replication configuration.

Create LDEVs within the Management Pool (Journal Vol, & Command Device)

This procedure assumes completion of dynamic pool creation during UCP Pro initial configuration in your environment.

To create volumes that will be used for Journal Volume and Command Device, follow these steps on BOTH sites.

1 From the Actions menu, point to Logical Device and then click Create LDEVs. The Create LDEVs window opens. 2 From the Provisioning Type list, click Dynamic Provisioning. 3 From the Emulation Type list, click OPEN-V. 4 From the RAID Level list, click RAID-6 (6D+2P). These options allow you to filter the available RAID group volumes. 5 Select the pool options.

Click Select Pool. The Select Pool window opens.

Select the pool that represents the Management Pool (Pool 0) in the Available Pools area and click OK.

Type the capacity in LDEV Capacity and click a unit of measure from the list. This solution uses 5GB LDEVs for the command device..

Type the number of LDEVs of the size in LDEV Capacity to create in Number of LDEVs.

In the LDEV Name area, type a prefix in Prefix and type an initial number in Initial Number. 6 Add the LDEVS.

Click Add. The Selected LDEVs pane populates.

Click Next.

Click Finish. The confirmation window for creating LDEVs opens.

Click Apply. 7 Repeat steps 1-6 for Journal Volume. Capacity for Journal volume in this solution is 50GB.

Add Volumes to Management Node HSDs

To add volumes to the management hosts that will be used for as Command Device, follow these steps on BOTH sites.

1 From the Actions menu, point to Logical Device and then click Add LUN Paths. The Add LUN Paths window opens. 2 From the Available LDEVs list, select the 5GB volume that will become the command device, and click the Add Button to

add it to the selected LDEVs list. 3 Click the Next button. 4 From the Available Host Groups list, select all 8 Management HSDs on ports (one for each Mgmt Host on CL1-A, CL2-A,

CL3-A, and CL4-A). 5 Click Add button to add Host Groups to the Selected Host Group list. 6 Click Next Button 7 Verify LUN paths, and click Finish button.

Create a Command Device

Create command devices at the local and remote sites. Verify that all the command devices are accessible by the ESXi hosts through multiple I/O paths.

To create a command device using Hitachi Storage Navigator, follow these steps:

1 Create an LDEV in the Hitachi Dynamic Provisioning pool as described in "Create LDEVs within a Dynamic Provisioning Pool " section and map to the created host groups.

2 Convert an LDEV to a command device in Storage Navigator by following these steps: 1 Select the LDEV from the Explorer > Logical Devices area.

17

2 From the Actions menu, point to Logical Device and then click Edit Command Devices. 3 In the Edit Command Devices window, click the Enable option. 4 Click Finish. 5 Click Apply.

UCP Pro for VMware vSphere SRM environment

This solution is built on a VMware HA model for both the UCP Management nodes, as well as the compute nodes. The use of multiple ESXi hosts placed in a VMware HA cluster adds an additional layer of protection and redundancy for Virtual Machines in a disaster recovery situation. The UCP management nodes are configured from the factory and during the onsite configuration as depicted in the figure below.

Management Cluster Site B

VSP SiteA VSP SiteB

SiteAESXi01

SiteAESXi02

SiteBESXi01

SiteBESXi02

Boot Boot

00:FD:03ManagementDatastore1


Boot Boot

00:FD:04Datastore2

Management Cluster Site A

AD VM

HCS VM

SQL VM

UCP VM

Util VM

VC5 VM

AD VM

HCS VM

SQL VM

UCP VM

Util VM

VC5 VM

00:FD:03Datastore1

Figure 7 - Management Nodes

For this SRM/SRA solution, an additional set of components are required to be configured in the management cluster, and in the pool for the management volumes on both sites.

1 Create virtual machine for SRM in cluster (Windows 2008 R2) 2 Command device created in management pool (see Create LDEVs within the Management Pool & Create a Command

Device above. 3 Command device attached to both management nodes (see Add Volumes to Management Node HSDs above). 4 Present the Command Device to the SRM VM as an RDM (Raw Device Mapping) volume. 5 Install SRM 5.1 on SRM VM

18

6 Install Hitachi SRA 2.01.3 on SRM VM 7 Install Hitachi Command Control Interface (CCI) on SRM VM

Once completed, the management cluster will look like the following figure:

Management Cluster Site B

VSP SiteA VSP SiteB

SiteAESXi01

SiteAESXi02

SiteBESXi01

SiteBESXi02

Boot Boot



Boot Boot

00:FD:04Datastore2

00:FD:05CMD Dev

(as RDM)

00:FD:05CMD Dev(as RDM)

Management Cluster Site A

AD VM

HCS VM

SQL VM

UCP VM

SRM VM

Util VM

VC5 VM

AD VM

HCS VM

SQL VM

UCP VM

SRM VM

Util VM

VC5 VM

00:FD:03Datastore1

Figure 8 - SRM VM w/Command Device

Looking at both site configurations with the Management environment, in conjunction with the compute environment and the SRM protection group, the UCP Pro / SRM solution will look like the following:

ucp srm whitepaper

Documents