Joined: 26 Apr 2004 Posts: 4650 Location: Raleigh, NC, USA
I would think that you would either:
A. Write a program that reads the control file (with the input/out/extract dataset names), and then builds one 100-step job with all of the proper dataset allocations, and then submits it.
B. Have the job (JOBA) read the control file and build the JCL for the next job (JOBB), filling in the proper dataset allocations, then submitting JOBB. JOBB would run and, if successful, remove the one line from the control file, and submit JOBA. This process would continue until either: a. there are no more entries in the conrol file or b. one of the JOBB jobs abends.
C. Have the program that creates the output and extract datasets use a dynamic allocation routine for the DD names as specified by the control
D. Run a TSO Batch job to do all of this with CLIST or REXX and do the same basic process as C.