An allocation rule is selected by the analysis function to control this process. Selection of the allocation rule to be used is described in Creating statistics data sets.
Allocation rule parameters
An allocation rule contains user-supplied information that is based on values which are specified for the keywords discussed in this section.
Each allocation rule contains a data set name mask that is used to build the name of the statistics data set.
The mask consists of a string of literal characters and substitution parameters. The following table lists the substitution parameters that are available for building a statistics data set name:
Three-digit area number
Four-digit area number
Five-digit area number
Current date ('Dyyddd')
Current time ('Thhmmss')
PROC STEP name
The purpose of the data set name mask and substitution parameters is to generate a unique name for each statistics data set. A good technique for creating unique names is to use &DATE and &TIME substitution parameters as part of the data set name.
Generation data group
You can also generate a unique statistics data set name by using generation data groups (GDGs).
Use the data set name mask to construct the GDG base name, but do not include relative or absolute generation numbers. The analysis function automatically appends the next generation number to the data set name if it detects that the name is a GDG base name.
NULLFILE as data set name
The generated data set name has a special significance when the first eight characters of the generated data set name is NULLFILE. When this data set name is specified, no statistics data set is created. The statistics catalog entry is still added, however, to identify that analysis was performed for the area and to store the summary statistics that are gathered during analysis.
When no statistics data set is created, or if the statistics data set has been deleted or is not available, only the summary statistics that are stored in the statistics catalog entry are available for subsequent processing. The summary statistics are not sufficient for regeneration of all analysis reports, but can be used to generate the following information:
Area Summary Report
Abbreviated version of the Free Space Analysis Report
Information stored in the History File (For more information, see the .)
Each allocation rule contains a unit name that specifies the device type where the new statistics data set is to be stored.
EXPDT and RETPD
Each allocation rule contains an expiration date (EXPDT) or a retention period (RETPD) to be associated with the new statistics data set when it is allocated.
DATACLAS, MGMTCLAS, and STORCLAS
Each allocation rule contains any combination of SMS data class, management class, and storage class.
Each allocation rule contains a volume serial number (VOLSER) to specify the storage volume where the new statistics data set will be placed.
Default allocation rule
When a repository catalog is initialized, a default global allocation rule is inserted. This default rule specifies the DSNAME parameter as NULLFILE, and the UNIT parameter as SYSDA. By default, no statistics data set is created for any DEDB area. However, the statistics catalog entries containing the summary statistics are still created.
You might want to modify the allocation rule (or rules) so that the analysis function will create the more detailed statistics data sets. To set up your own allocation rules, see Creating and modifying allocation rules.