Previous Topic: Multivolume Output ProcessingNext Topic: Uncataloged Data Sets


Multiple File Processing to Virtual Volumes

Multiple file processing or stacking to Virtual Volumes is supported, but not normally needed for application data sets. Media is not wasted since CA Vtape provides automatic stacking of the Virtual Volumes on Backstore Tapes.

Stacking to Virtual Volumes make sense when running backup jobs which write a data set for each data set or table being backed up. A job backing up 100 data sets or tables writing one data set or table per Virtual Volume would need to have 100 cache LDSs defined, 100 scratch VOLSERs selected, 100 VOLSERs updated in the Global VCAT, and so on. If the data sets or tables would fit on five full Virtual Volumes if stacked, the time spent performing these tasks would be reduced by 95%. These types of jobs will run faster when stacking versus when they are not stacking.

CA Vtape limits the number of files per Virtual Volume to 256 for 400 MB and 800 MB Virtual Volumes to avoid recall performance and recovery implications. When file 256 is written, an end of volume condition is presented by CA Vtape and a volume switch is done. The higher capacity Virtual Volumes have a limit of 9999 files per Virtual Volume.

If more than one data set is written to a Virtual Volume, all of them will assume the same Externalization group as the first data set on the volume. This occurs for the second and subsequent data sets regardless of the Externalization group assigned by the Data Set Filter List or Data Class Filter List. The subgroup for all the files residing in the same Virtual Volume is determined by information returned by the tape management system or the file with the longest retention period within the first 32 files.