Specifying small data set packing

The SMALLDATASETPACKING parameter of the SETSYS command controls the sending of data sets to the SDSP data sets.

In specifying small data set packing, you specify the actual physical size of the data sets to be considered as candidates for migration to an SDSP data set. You can specify sizes as large as 800 KB. A candidate data set is eligible for migration to an SDSP data set if its estimated size, after compaction, is less than or equal to 400 KB or to the value you specify, whichever is smaller. The compaction size is estimated using an internal default of 50%. If the data set is not compacted, it is a candidate for migration to an SDSP if its actual size is less than or equal to 400 KB or to the value you specify, whichever is smaller.

Small data set packing causes data sets that are smaller than the size specified to be written as records in special VSAM data sets that you have allocated on the ML1 volumes. Because the minimum allocation unit is a track, writing small data sets as records in a larger data set results in saving space on the ML1 volumes.

The number of SDSP data sets defined must be at least equal to the maximum number of concurrent volume migration tasks that could be executing in your complex.

Example: You can add the following command, the command to be added to the ARCCMDxx member in each DFSMShsm host:
SETSYS SMALLDATASETPACKING(KB(110))
Recommendation: Configure recall and ABARS use more SDSP data sets in case some data sets become full during migration.

Related reading

For more information about small data set packing, see Managing SDSP data sets.