Regardless of the conversion method chosen, Quick or Phased, some steps are common to both. These common steps form the minimum requirements that must be considered so that dynamic mode Virtual Volumes are properly allocated and can be mounted by any SVTS subsystem. The common items include the following:
A complete description and example of the data class and storage class definitions required for CA Vtape can be found in the section Setting up Dynamic Cache Management.
Important! The data class and storage class attribute definitions must allow the SVTS subsystem to allocate DFSMS managed linear data sets (LDS) that are single striped extended format, (multi-striped data sets will prevent CA Vtape from releasing unused DASD space) that can be allocated as multivolume and that support Space Constraint Relief.
Note: Assigning available DASD volumes to the dynamic cache storage group to support the conversion of the first SVTS subsystem, rather than converting static cache DASD volumes, will eliminate any conversion impact on the subsystems that continue to run in static mode. These subsystems will continue to have the full static cache available to support their workload until DASD volumes are converted to support the conversion of the second SVTS subsystem to dynamic mode.
Note: If you have more than one CA Vtape subsystem and each has its own set of parmlib members, you must make changes in more than one parmlib member.
Use the SVTS DISPLAY CACHE command to check the data and storage class values assigned by DFSMS. This command also displays DASD storage usage and gives you some sense of useable space within the storage class. Remember that class constructs specified in CA Vtape may be inadvertently overridden by the automated class selection (ACS) logic. Check to ensure that the construct values assigned by SVTS are assigned and not somehow overridden.
|
Copyright © 2013 CA Technologies.
All rights reserved.
|
|