Collecting performance data, Performance data – HP XP P9000 Performance Advisor Software User Manual
Page 71
Collecting performance data
After completing the configuration data collection for the XP and the P9000 disk arrays, schedule the
performance data collection for the associated components, which belong to the following component
types:
•
DKC
•
Ports
•
RAID Groups
•
Ext RAID Groups
•
THP pools
•
Snapshot pools
•
Cont. Access Journals
You can create two performance data collection schedules for an XP or a P9000 disk array, as it
enables you to frequently monitor the respective components. The components that are not selected
as part of the first schedule are automatically added to the second schedule that you create. You can
set different collection frequencies to collect data for the DKC, ports, and the RAID groups. You can
also enable the schedules for automatic updates. The collection frequency set for a RAID group in a
schedule is applicable for any component of the continuous access journal, snapshot, ThP, and the
external RAID group types. It is applicable only if they are selected for performance data collection.
The following are important notes on the performance data collection:
•
The performance data collection does not start if the configuration data collection is in progress.
After the configuration collection is complete, the performance data collection starts automatically.
•
While creating a performance data collection schedule, you can select a command device that
is mapped through two different ports.
•
In a multipathing environment, ensure that a command device is not exposed to a host from two
different ports. It stops the current performance data collection, as the schedule configured to obtain
the collection is automatically deleted.
•
HP recommends that you set the data collection rate to one hour or less because of management
station performance and field rollover. P9000 Performance Advisor collects performance data on
all LDEVs in the XP and the P9000 disk arrays that communicate with the management station
through their respective hosts. The hosts that display their status as Received under the Host Inform-
ation tab constitute the superset of the mapped LDEVs. The performance data collection is not
limited to the number of LDEVs that the host station is mapped to use. If you set the collection rate
too narrow (less than 5 minutes), it results in reduced responsiveness from the management station.
•
If you split the external RAID groups, RAID groups, ThP pools, and the snapshots into two schedules,
you cannot see the Total I/O, Total MB/s, and the Total Tracks fields in the LDEV table.
•
If you split the journal LDEVs, external RAID groups, RAID groups, ThP pools, and the snapshots
into two schedules, and in charts, you select All from the LDEV drop-down box, the combined data
points from both the schedules are plotted on the chart. In addition, repeated time stamps are
displayed if the collection frequency for both the schedules is the same. As a result, incorrect values
are plotted on the graph.
For more information about charts, see “
•
If you are collecting the performance data for the first time, it takes longer than usual for P9000
Performance Advisor to collect the data. The subsequent performance data collection is as per the
time specified in the Frequency section.
•
If you have performed a one-time configuration data collection and later the XP or the P9000 disk
array configuration is modified, HP recommends that you perform a fresh configuration data col-
lection, so that the performance data collected is for the latest configuration.
HP StorageWorks P9000 Performance Advisor Software User Guide
71