Running Batch Jobs

There are several batch jobs that are part of the CMA process, especially the import step. And in some cases, a single batch jobs may process multiple states in the same business object lifecycle. Implementations must decide the best way to manage the batch job submission depending on how they plan to work.

Note that for the CMA batch processes, especially the “apply” batch processes, the setting of the Thread Pool Name parameter is important. This parameter specifies the thread pool that will be executed by the batch. Since the CMA apply processes update administrative data that is usually read from a cache during batch processing, the thread pool named here must be one with the L2 cache disabled. In addition, after successfully applying the migrated data, the L2 cache of all thread pools must be refreshed. For more information, see Caching Considerations.

CAUTION:
Be sure that the Thread Count set when submitting the batch job does not exceed the number supported by the thread pool. Otherwise the extra threads will simply wait until the supported number of threads are finished.

Refer to the parameter descriptions in the batch control metadata for more information about filling in the parameters.

For additional details on submission controls, refer to the topic Batch Job Submission - Main in the Batch Jobs section.