2. PERFORMANCE REPORTING ANALYSIS › 2.10 Product Measured Usage Analysis › 2.10.2 Usage Guidelines › 2.10.2.2 Combined Hourly System Usage by Product
2.10.2.2 Combined Hourly System Usage by Product
The Combined Hourly System Usage by Product reports analyze
information from the Measured Usage Global (BATMUG) file,
merged with the Hardware Utilization (HARCPU) file. The
first two BATMUG Frequency Reports will provide a software
inventory of any products reported with the SMF type 89
(Usage Data) records along with a list of vendors represented
in the data. The last BATMUG Frequency Report provides a
matrix of products versus the LPARs on which they ran. Use
the information provided in these reports to subset the input
data, reducing the size and scope of the reports.
To further subset the input data beyond the data selection
options available on the query composition panel, you can
specify a global data selection exit by coding a SELECT macro
under the Extended Options panel. If you enter Y (Yes) in
the Specify global data selection exit field, the Global Data
Selection Exit Panel, which is shown in Figure 2-228, is displayed.
This panel lets you enter SAS statements that will
perform special data selection functions.
The SELECT MACRO can be used for special selection code based
on any data element in the BATMUG file. A typical use may be
to limit data analysis by excluding a specific product from
analysis. The following example will exclude all releases of
z/OS from the analysis.
/-------------------------------- SAS Statements -------------------------------\
| Command ===> Scroll ===> CSR |
| |
| Inquiry Step: Combined Hourly System Usage Analysis |
| |
| Line Cmds: I Insert D Delete R Repeat M Move C Copy |
| |
| Cmd SAS Statements for: Specifying global data selection |
| - ------------------------------------------------------------------------ |
| _ - %MACRO SELECT; |
| _ IF MUPROD NE 'z/OS'; ___________________________________________________ |
| _ ________________________________________________________________________ |
| _ - %MEND SELECT; |
| ****************************** BOTTOM OF DATA ******************************** |
| |
\---------------------------------------------------------------------------------/
Figure 2-228. Global Data Selection Exit Panel
Because IBM defines its billing cycle as the second day of
one month to the first day of the next month, when generating
a software inventory for the purpose of subcapacity pricing,
the query must be set up to report on data from that time
period only.
The All-Hour report will do the following:
o Determine the LPARs on which each product executed.
o Calculate and sum the product's MSU rate for each hour and
LPAR in which it ran.
o Sum the LPARs' four-hour rolling averages across all LPARs
on which the product ran.
o Generate a report section for each hour the product ran.
o Display the combined systems' four-hour rolling average
along with the combined products' MSU rate.
Subcapacity pricing is based on the concept of the highest
observed rolling four-hour average utilization of the LPAR or
LPARs where a product runs. The simultaneous combined
utilization of these LPARs is determined for each hour in the
reporting period and the highest observed combined
utilization is used as the basis for pricing the product.
The High-Low Combined System Usage report is simply a subset
of the All-Hour Combined System Usage Report. It is designed
to display the four highest observed combined utilization
hours and the four lowest for each hour the product ran.
Using this information you can begin to study the possible
effects of workload shifting in respect to IBM's subcapacity
pricing methodology. There are many advantages and
disadvantages to moving to a subcapacity pricing method, so
careful planning and analysis is important.
