IBM Mainframe Forum

That's an awful lot of data to chug through with REXX.

I'll leave with this. For COBOL search for any of the known methods of performing a dynamic allocation (BPXWDYN, PUTENC ,or IKJTSOEV and IKJEFTSR). Or, use your SORT to read through the data, find the record number where each new policy number starts, and have it generate a series of appropriate SORT control statements and the necessary DD statements for inclusion in a subsequent job, or optionally create and submit a whole new job.

COBOL actually has two ways to do dynamic allocation now: using environment variables or calling BPXWDYN. Environment variables aren't quite as flexible as BPXWDYN and may not provide you all the options you need for an output file. If you Google BPXWDYN, you can find plenty of documentation on it and how to use it. BPXWDYN is documented in the Unix System Services bookshelf manual Using REXX and z/OS UNIX System Services manual.