3. REPORTS › 3.2 MICF Inquiries › 3.2.4 Post Processor CPU Activity
3.2.4 Post Processor CPU Activity
The CPU Activity report provides information about each
processor on a given SYSID as well as on system address space
activity. Additional reports are provided for PR/SM
environments, including a Partition Data report, an LPAR
Cluster Report, and a Group Capacity Report.
INTENDED USE:
The CPU Activity section of the report helps with the
analysis of processor utilization, including busy and wait
times, percentages busy, and I/O interrupt rate. The system
address space analysis can be used to detect problems such as
a backlog of address spaces waiting for access to a
processor.
The Partition Data report produced for a PR/SM LPAR
environment can be used to examine the partition
configuration and dispatching activity. Moreover, it allows
data centers to have a better understanding of their true
resource utilization by showing the resource consumptions
separately, by type of physical processor assigned to each
partition. The type can be general purpose processor (CP),
internal coupling facility (ICF), integrated facility for
Linux (IFL), System z Application Assist Processor (zAAP),
or System z Integrated Information Processor (zIIP).
Finally, separate total lines are provided for each pool of
processors.
For logical partitions managed by WLM, additional information
related to the Intelligent Resource Director (IRD) is
provided in both the Partition Data and the LPAR Cluster
sections of the report.
The Group Capacity section of the report provides all
information to understand how the CPU resources (MSUs) are
consumed in the various capacity groups.
See chapter 6 for more information about total system
utilization percentages in configurations where multiple
processor types are used.
REPORT FORMAT:
See section 3.1.1.1 for an example of the output generated by
the CPU Activity report.
DATA SOURCE (file/timespan):
HARCPU, HARIPU, HARLPC and HARVPA (PR/SM LPAR systems only)
at the DETAIL, DAYS, WEEKS, or MONTHS timespan.
DATA ELEMENTS USED AND CALCULATIONS:
Consult the SAS source code used to generate this report in
sharedprefix.MICS.SOURCE member:
o @RMFCPU4 - for all systems
HOW TO PRODUCE THE REPORT:
o Enter the CA MICS Workstation Facility (MWF).
o Choose option 2, CA MICS Information Center Facility
(MICF).
o Choose option 2, Database Inquiries.
o Select RMFLPA, whose Catalog Group is PERFORM.
o Complete the MICF Execution panel. (If you are not
familiar with MICF, see the MICF User Guide and the MICF
Reference Guide.)
o Complete the specification panel illustrated below.
(Descriptions of the fields follow the panel.)
+----------------------------------------------------------------------+
|--------------------- Post Processor CPU Activity Report -----------|
| Command ===> |
| |
| Composing Inquiry: RMFLPA - Post Processor CPU Activity Rpt |
| Database ===> ___________ |
| Cycle(s) ===> __ - __ |
| Timespan ===> ______ (DETAIL/DAYS/WEEKS/MONTHS) |
| |
| Data Selection: |
| dd mmm yy dd mmm yy dd mmm yy dd mmm yy |
| Date ===> __ ___ __ - __ ___ __ __ ___ __ - __ ___ __ |
| Hour ===> __ - __ |
| Zone ===> _ _ _ _ _ _ _ _ _ |
| Sysid ===> ____ ____ ____ ____ ____ ____ ____ ____ ____ ____ |
| |
| Summarization Break Variable ===> ________ |
| Summarization Technique ===> _________ (DETAIL or COMPOSITE) |
| |
| SAS Subsetting Statements ===> |
| _______________________________________________________________ |
+----------------------------------------------------------------------+
Database - Required. No default.
Specify up to six database unit IDs (the IDs of the
CA MICS database units from which the files are read).
If CA MICS files exist in only one database unit, this
parameter is set automatically. To display a selection
list of valid database unit IDs, specify ? (question
mark).
Cycle(s) - Required. Defaults to 01.
00-99 in either data entry field; leading zeros are
inserted automatically. The CA MICS files cycle
(specify first field) or cycle range (specify both
fields) for the inquiry execution.
Timespan - Required. Defaults to DAYS.
The timespan of the CA MICS database from which the
primary input files are selected. See Chapter 3 for a
listing of file requirements by report. The CA MICS
required input files must be available at the specified
timespan.
Date - Optional. No default.
Up to two date ranges in DD MMM YY format. The first
value in each date range is the starting date and the
last value is the ending date. The ending date must be
higher than the starting date. The date ranges should
not overlap, but you can specify the higher range
before the lower range. An input observation is
selected if its date value is greater than or equal to
the starting date and is less than or equal to the
ending date of any specified date range.
Hour - Optional. No default.
An hour range from 0 to 23. The first value is the
starting hour and the last value is the ending hour.
The ending hour must be higher than the starting hour.
An input observation is selected if its hour value is
greater than or equal to the starting hour and is less
than or equal to the ending hour of the specified hour
range.
Zone - Optional. No default.
One-character zone value from 1 to 9. You can specify
up to nine zone values. An input observation is
selected if its zone value is equal to any of the
specified values.
SYSID - Optional. No default.
1-4 character SYSID value. You can specify up to ten
SYSIDs. An input observation is selected if its SYSID
value is equal to any of the specified values.
Summarization Break Variable - Optional. No default.
Data element name to be used as control-break variable.
Valid entries are STARTTS, HOUR, ZONE, DAY, WEEK,
MONTH, YEAR, and SYSID. CA MICS summarization
facilities are used to preprocess the files used by the
report programs. These facilities allow that a data
element be defined as a control-break variable, and
that the summarized file(s) are sorted so the break
variable is last in the sort key list. A single
summarized observation is then written from multiple
observations, with the same break variable value,
whenever the break variable value changes.
Summarization Technique - Required. Defaults to DETAIL.
Specify DETAIL to cause the input files to be processed
and reported in date/time order, or COMPOSITE to force
ALL occurrences of the same value of the summarization
break variable to be summarized and reported together.
See section 3.1.2.1.4 for a more detailed explanation
of the summarization control techniques.
SAS Subsetting Statements - Optional. No default.
Free-form user-supplied SAS code. During the creation
of the work files used to generate the reports,
user-defined SAS code may be included to modify or
subset the data.
IMPORTANT: No validity checking is done on the
user-defined code. If it contains syntax or logic
errors, the inquiry will either terminate abnormally or
will produce unpredictable results.
Examples of valid statements:
IF SYSID EQ "SYS1";
IF DEVTYPE EQ: "3390";
USAGE CONSIDERATIONS:
This inquiry provides a user exit that you can use to perform
additional data selection from the CA MICS database. To
invoke this exit, you must first copy the shared inquiry into
your private MICF catalog, then select it for modification.
Finally code SAS statements in the SELECT macro, bracketed by
the %MACRO and %MEND statements.
The default for the SELECT macro is null. This macro is
invoked when the observations are read from the specified
CA MICS file(s). You can use it to delete observations from
the data as it is selected. For example, in a system
activity study, you might want to exclude test systems
starting with a T from the analysis. You would then code
the following:
%MACRO SELECT;
IF SYSID NE: 'T';
%MEND SELECT;
Note that you do not have to code the %MACRO and %MEND
statements, as they are already provided for you.
IMPORTANT: You must code each specified SAS statement in
accordance with the syntax of the SAS Macro Language, as
described in the SAS manuals. No validity checking is done
on the defined code. If the user code contains syntax or
logic errors, the inquiry will either terminate abnormally or
will produce unpredictable results.
************************************************************
** Note: This report cannot be produced if the CA MICS **
** database contains essential data elements only.**
************************************************************
