Previous Topic: COPY Command SyntaxNext Topic: SMS and MOVE/COPY

MOVE/COPY › VSAM Cluster Considerations

VSAM Cluster Considerations

CA Disk processes all components of VSAM clusters selected for processing. When any VSAM component name is encountered during the scan of a volume, and selection criteria causes it to be processed, it and all of its related components will be moved or copied. Related components will be moved even if they exist on volumes other than the source volume.

Move and copy processing runs above the 16meg line for VSAM data sets.

The movement of data is accomplished by two methods, control interval method and IDCAMS service.

Control Interval method is done via VSAM IO macros. After the data is moved, the target data set is a mirror image of the source data set. note that the control interval method SHOULD NOT be used to move the data to a different device type because the image of the target data set cannot be the same as the source data set. The data is moved using the control interval method in the following cases:

  1. The data set is Linear or RRDS data set.
  2. The data set is ESDS and the record count fields in the statistic are zero.
  3. The data set is ESDS, the record count fields in the statistic are not zero, and the value of sysparm 'VSARCFMT' is 'C'.

IDCAMS moves data logically and any free space in the control interval is reconciled. In the event of an error, you can see IDCAMS error messages in your CA Disk report. These messages will supplement error messages issued by CA Disk.

The data is moved using IDCAMS services in the following cases:

  1. The data set is KSDS.
  2. The data set is ESDS and the record count fields in the statistic are not zero. The sysparm 'VSARCFMT' can be used to force the data to be transferred using the control interval method.

The following special considerations apply to usage of the target volume list when processing VSAM clusters:

  1. When a VSAM cluster's data and index components are moved from the same volume, they will be moved to a single target volume. When they exist on different source volumes, CA Disk will attempt to place them on different target volumes.
  2. Alternate index components are allocated separately from data and index components. When multiple target volumes are specified, the alternate index component can be placed on the same volume as the base cluster, or on a different volume, depending on the results of the allocation algorithm.
  3. System Resource Manager (SRM) routines select the volume to be used for the target allocation from the target volumes.
  4. An attempt will be made to avoid placing cluster components on volumes that have less free space than specified on the VATHRESH parameter. If this is not possible, allocation will choose the volume with the largest free space that can contain the cluster.
  5. If a data set is selected without any volume selection criteria, CA Disk will only use the volumes specified in the TOVOL= parameter as candidate volumes. In other words, for a multi-volume cluster, the cluster can end up using one or more of its original volumes. To force CA Disk to use only NEW volumes, include the NEWVOLUMES parameter described in NEWVOLUMES.