Specifies for the Full Analysis Report, whether to include the rule's conditions and evidence in the report output. By default, the Performance Manager uses the /EXPLAIN qualifier for a batch process or if /OUTPUT is specified or asks you if the process is interactive.
Specifies the maximum number of hot files to list per disk volume in the Hotfile Statistics section of the Performance Evaluation Report. By default the maximum number of hot files is 20.
Specifies which sections of the Performance Evaluation Report are to be included. The negatable options are as follows:
By default, the Performance Manager includes all sections except SCS statistics. Rule statistics are only available with archived data.
Specifies the preparation of a report for only one of the selected nodes. By default, the Performance Manager prepares reports for all selected nodes. cluster-wide statistics are always included.
Specifies an output file specification as a destination for the report output. By default, the output file destination is SYS$OUTPUT.
The previous code lets you tailor the process statistics section of the Performance Evaluation report. You can specify the focus of the report to obtain slightly different sorts of statistics that pertain to the focus area. The grouping, merging, and sorting of the process data is controlled with the primary and secondary key settings. To use a given primary or secondary key, you must have previously specified the process key level with the SELECT command. See the description of the PERFORMANCE_EVALUATION option with SELECT. Also, you can specify whether a cluster-wide report, or a node-by-node presentation is desired.
By default, the focus area is TRADITIONAL, that being an image based report showing relative resource consumption. Reports are provided by node, unless otherwise specified.
Where:
|
FOCUS Types: |
Provide: |
|---|---|
|
TRADITIONAL |
An 80 column report showing process CPU, memory and IO statistics on a per image activation basis, or as a relative percentage. The default primary and secondary keys are MODE and IMAGENAME. The SUMMARY focus report is also provided with the TRADITIONAL flavor. |
|
SUMMARY |
An 80 column report showing process CPU, memory and IO counts on a per image activation basis. The default primary key is MODE (no secondary key). |
|
GENERAL |
A 132 column report showing process CPU, memory and IO statistics primarily as rates. Some UAF parameters are also provided. The default primary and secondary keys are USERNAME and IMAGENAME. |
|
MEMORY_RELATED |
A 132 column report showing primarily process memory related statistics. Some UAF parameters are also provided. The default primary and secondary keys are IMAGENAME and USERNAME. |
|
IO_RELATED |
A 132 column report showing primarily process IO related statistics as rates. The default primary and secondary keys are USERNAME and IMAGENAME. |
|
CPU_RELATED |
A 132 column report showing primarily process CPU related statistics. The default primary and secondary keys are USERNAME and IMAGENAME. |
|
Primary Keys: |
Provide: |
|---|---|
|
MODE |
Group process statistics by the process mode (Interactive, Batch, Network, or Detached). |
|
USERNAME |
Group process statistics by the process's User name. |
|
IMAGENAME |
Group process statistics by the process's Image Name. |
|
UIC_GROUP |
Group process statistics by the process's UIC Group. |
|
PROCESS_NAME |
Group process statistics by the process name. |
|
WORKLOAD_NAME |
Group process statistics by the workload name. You must specify /CLASSIFY_BY to indicate the workload family that you intend to use. |
|
ACCOUNT_NAME |
Group process statistics by the process's account name. |
|
PID |
Group process statistics by the process's PID. |
|
Secondary Keys: |
Provide: |
|---|---|
|
MODE |
Process statistics detail lines by the process mode (Interactive, Batch, Network, or Detached). |
|
USERNAME |
Process statistics detail lines by the process's User name. |
|
IMAGENAME |
Process statistics detail lines by the process's Image name. |
|
UIC_GROUP |
Process statistics detail lines by the process's UIC Group. |
|
PROCESS_NAME |
Process statistics detail lines by the process name. |
|
WORKLOAD_NAME |
Process statistics detail lines by the workload name. You must specify /CLASSIFY_BY to indicate the workload family that you intend to use. |
|
ACCOUNT_NAME |
Process statistics detail lines by the process's account name. |
|
PID |
Process statistics detail lines by the process's PID. |
|
Other Options: |
Provide: |
|---|---|
|
[NO]CLUSTER |
A summary of process data for the entire cluster, scaled by CPU speed. The default is NOCLUSTER |
|
[NO]BY_NODE |
Per node detail of cluster process data. The default is NOBY_NODE if you specify CLUSTER. |
If the value for the secondary key is the same as the primary key, no secondary level breakout occurs. This also happens if you specify the primary key and no secondary key is given.
The CLUSTER and BY_NODE keywords allow you to specify that the process statistics section of the Performance Evaluation Report is to present data combining process information for all selected nodes (CLUSTER), and if so, whether the (BY_NODE) detail should also be included. By default, process data is not combined for all selected nodes.
When the CLUSTER option is used, the percentage of CPU Utilization for the data line of each process is scaled according to the processor speeds of the nodes in the cluster. The speed ratings can be changed using an auxiliary knowledge base.
If you specify CLUSTER or BY_NODE, a by-node breakdown of the process data is provided following the line representing the cluster-wide data.
Specifies which sections of the Tabular Report should be displayed. By default, all are displayed.
For node analysis, the available sections are as follows:
For PROCESS_STATISTICS and EXTENDED_PROCESS_STATISTICS, the process data must be selected with the PROCESS=ALL keyword. For example:
PSPA> SELECT=PERFORMANCE=PROCESS=ALL
For cluster analysis, the available sections are as follows:
|
Copyright © 2008 CA.
All rights reserved.
|
|