WebSphere Extended Deployment Compute Grid, Version 6.1.1
             Operating Systems: AIX, HP-UX, Linux, Solaris, Windows, z/OS


Other considerations when running the parallel job manager

There are other considerations that help you understand how to optimally use the parallel job manager.

Transaction timeouts

You can keep the default value for the property TransactionTimeOut in the top-level job xJCL, or you can adjust it depending on the transaction timeout of the sub job.

Job logs

You can view job logs for a sub-job from the job management console. The PJM retrieves sub-job logs for its logical job and aggregates them into its top-level job log.

Failover scenarios

If the top-level job and several sub-jobs are in the restartable state, then only restart the top-level job. If you restart any of the sub-jobs manually, then the top-level job will not process the logical transaction properly.

Disaster recovery

You can use the jobrecovery script when a primary site fails to allow a secondary site to take over. See the jobrecovery.bat|.sh script .




Related concepts
Managing Compute Grid jobs and their environment
Related tasks
Sample application for the parallel job manager
Related reference
System Programming Interfaces (SPI) and properties
parallelJobManager.py script
Concept topic    

Terms of Use | Feedback

Last updated: Oct 30, 2009 1:38:02 PM EDT
http://publib.boulder.ibm.com/infocenter/wxdinfo/v6r1m1/index.jsp?topic=/com.ibm.websphere.gridmgr.doc/info/scheduler/ccgpjmother.html