Migrating Batch Data

The migration with data movement strategy depends on being able to identify the data sets for a specific application by the data set name. In the sample job, the HLQ is used to identify the data set.

The data is moved to the PRIMExx and LARGExx storage groups, based on the size of the data sets using DFSMSdss COPY and the source volumes that the application's data resides on as another filter. Figure 1 shows sample JCL that you can use to migrate the first application's data to system management. The sample job uses the following parameters:

The LOGINDYNAM(D65DM2,D65DM3) parameter tells DFSMSdss to move all data sets on volumes D65DM2 and D65DM3.
The DATASET(INCLUDE(1STAPPL.**)) indicates that, for the volumes selected, only data sets having the 1STAPPL HLQ are copied.

After you successfully migrate the first, modify the data set and volume filters to migrate other applications.

You cannot migrate to striped data sets with DFSMSdss. You can exclude these data sets from the initial COPY and then recreate them as striped data sets through the use of standard z/OS® utilities, such as access method services or IEBGENER.

Figure 1. DFSMSdss Job Migrating Batch Data to System Management

    COPY DATASET(INCLUDE(1STAPPL.**)) -
         LOGINDYNAM(D65DM2,D65DM3)              -
         ALLDATA(*)                             -
         ALLEXCP                                -
         CATALOG                                -
         SPHERE                                 -
         DELETE                                 -
         PURGE                                  -
         TGTALLOC(SOURCE)                       -
         TOLERATE(IOERROR)                      -
         WAIT(2,2)

If you temporarily remove data sets for a batch application from system management, you can still use the DATACLAS parameter to supply the data set characteristics. Data classes apply to both system-managed and non-system-managed data sets.