HP P6000 Replication Solutions Manager Administrator Guide (T3680-96069, June 2012)

Ensuring job success
To ensure job success and optimize performance:
• Minimize the number of simultaneous jobs, even if they involve separate arrays.
• Avoid unnecessary use of HP P6000 Command View when jobs are running.
• Avoid configuring storage and hosts while jobs are running. For example, do not change an
array configuration using HP P6000 Command View while replication manager jobs are
running. Resource changes can lead to job failures that require manual intervention to restore
resources to operational readiness.
• Avoid changing the network identification (computer network name or IP address) of a
management server or enabled host.
◦ Server identification change—If a management server identification is changed, and host
agents are not updated, jobs involving the enabled hosts will fail.
◦ Host identification change—If a host identification is changed' and affected replication
jobs are not updated, jobs will fail.
• Maintain network connections between the management server and enabled hosts, especially
while jobs are running, or jobs will fail.
• Coordinate planned shutdowns. Stopping a host causes running jobs to fail and prevents
scheduled jobs from starting.
• Be aware that parallel jobs that replicate the same volume group to the same target host are
not supported.
CAUTION: Failure to follow these guidelines can result in HP P6000 Command View errors.
Optimizing replication manager performance
This section describes replication best practices to optimize performance.
Snapshot and snapclone best practices
General snapshot and snapclone best practices:
• To create snapshots and snapclones:
Create a container before the snapshot or snapclone is needed.1.
2. Set the parent virtual disk to write-through cache state (flush data from cache of the virtual
disk to the physical media that represents the virtual disk).
3. Prepare the container for the snapshot or snapclone.
4. Within 60 seconds, attach the virtual disk parent to the container, thus creating the
snapshot or snapclone.
• Create and delete snapshots and snapclones during low-demand periods, or size the array
to meet performance demands during copy activities.
• Transition the source LUN to write-through cache mode before initiating the copy, and re-enable
write caching after the copy is initiated.
• Keep virtual disks as small as possible.
Snapclone best practices:
• Minimize the number of concurrent snapclone operations (use fewer virtual disks). Organize
clone operations into consistency groups of virtual disks, and then create snapclones of the
consistency groups sequentially.
Snapshot best practices:
Ensuring job success 33