HP StorageWorks Continuous Access XP user guide for the XP10000/XP12000 (December 2005)

86 Continuous Access XP for the XP10000/XP12000
Table 24 lists conditions affecting disk array performance and provides recommendations for addressing
these conditions.
Discontinuing Continuous Access XP operations
If you plan to use Continuous Access XP for nondisruptive data migration or duplication (see ”Using
Continuous Access XP for data migration and duplication” on page 103), configure and establish
Continuous Access XP operations, allow Continuous Access XP to synchronize volumes, redirect application
I/Os (if migrating), and discontinue Continuous Access XP operations. When you are ready to discontinue
Continuous Access XP operations, perform Continuous Access XP operations in the correct order to avoid
error messages. For example, Continuous Access XP will not delete an RCU path until you delete all
Table 24 Disk array performance conditions and recommendations
Condition Description Recommendations
Write-intensive
workloads
Write-intensive workloads, such as database
logging volumes, can have a significant
impact on disk array I/O response times.
Spread write-intensive data across several
volumes to minimize queuing. Also consider
increasing the duplex write line (DWL) of the
disk array using Performance Advisor
(Performance Advisor running at the same
time is not supported).
Large block size Workloads with large write block sizes, such
as DB deferred writes, can impact
performance.
Spread workloads with large write block
sizes across several volumes.
High host channel
demand
Demand on MCU’s host channels can affect
performance.
Spread workloads across several disk arrays
to utilize additional channels.
Sequential write
operations
Continuous Access XP operations can
have a negative impact on workloads with a
high percentage of sequential write
operations, such as batch processing
operations (for example, dump/restore, sort
operations).
Avoid performing restore operations to
volumes belonging to
Continuous Access
XP
pairs. Instead, restore data to a scratch
volume and then create the
Continuous
Access XP
volume pair.
Cache size
Large cache size improves read hit
performance, which allows more disk array
resources to be devoted to write operations.
Resulting performance improvements can
offset some or all performance losses due to
Continuous Access XP remote copy
operations.
Continuous Access XP Asynchronous
requires additional cache for sidefile data.
Insufficient cache resources can result in
command retries, state-change-pending
(SCP) notifications, and puncture conditions.
Consider increasing cache size of
Continuous Access XP disk arrays to
handle cache sidefile operations and
improve overall disk array performance. For
best results, cache and NVS capacity of
primary and secondary disk arrays should
be the same (for
Continuous Access XP
Asynchronous, RCU sidefile requirements are
twice that of the MCU) to enable the remote
site to function adequately during disaster
recovery.
RCU capacity Performance of RCUs directly affects
performance of MCUs. If an RCU is
overloaded with heavy update activity, MCU
and system performance can also be
degraded.
Distribute
Continuous Access XP remote
copy operations among several remote disk
arrays to avoid overloading any one RCU.
Paths
An inadequate number of paths may
decrease disk array performance. Performing
Continuous Access XP Synchronous
operations over long distances can also
degrade performance.
Continuous Access
XP
Asynchronous is recommended for long
distances.
Ensure an adequate number of paths
between the primary and secondary disk
arrays are installed. This is especially
important for disk arrays containing both
P-VOLs and S-VOLs.