HP StorageWorks P9000 Performance Advisor Software v5.3 User and Troubleshooting Guide (T1789-96322, August 2011)

Example (see Figure 25)
MP blade utilization by top
20 consumers
LDEV:0:18 (Backend)
A consumer's association with a processing type provides an understanding on the number of
processing cycles used by the consumer with different processing types.
For example, an LDEV 0:09 might be involved in processing frontend and backend requests. Its
processing type reveals whether the frontend or the backend requests have been high. Further to
the above-mentioned example:
• If the average utilization shown for the Frontend processing type is 30% and 25% is contributed
by the LDEV 0:09
• If the average utilization shown for the Backend processing type is 25% and 15% is contributed
by the LDEV 0:09
This data indicates that the LDEV 0:09 has been addressing high frontend requests when compared
to the backend activities.
NOTE:
The average utilization of an MP Blade is always on a scale of 1 - 100%.
Consumer type, Consumer ID,
and the processing type
1.37%
Indicates that 1.37% is the average MP blade utilization by the consumer 0:18 that belongs to the
LDEV category and associated with the Backend processing type.
It also indicates that out of the average MP blade utilization displayed for the Backend processing
type, 1.37% utilization is due to the consumer 0:18. For more information on processing types, see
“Viewing MP blade utilization by processing types” on page 306.
Percentage of MP blade
utilization by a consumer
IMPORTANT:
A consumer's percentage of MP blade utilization might or might not be constant for the entire duration
that you select. For example, it can be the utilization at a particular time stamp that is the maximum
compared to the utilization by other components during the selected duration.
Viewing MP blade utilization by processing types
IMPORTANT:
This section is applicable only for the P9000 disk arrays.
You can view the average MP blade utilization split up for the different processing types. For example,
if the average MP blade utilization by a Frontend processing type is significantly higher than the other
processing types, it indicates that more CPU cycles are utilized to process requests from the Frontend
consumers. Similarly, if the average utilization by a Backend processing type is high, it indicates that
more CPU cycles are utilized to process backend transfers for the Backend consumers.
Table 15 on page 306 lists the different processing types and their description.
Table 15 Processing types
DescriptionProcessing types
Indicates all the frontend activities involved in processing the I/O requests.Open-target
Using charts306