Yes, runnning in programs batch is always more efficient than in TSO - just as invoking assembler programs from REXX is more efficient than invoking REXX from assembler (via "ISPLINK" is it?) If JCL is 'strange', consider then native SMP/E. I never had a problem with JCL or with native SMP/E (and I ignored all IBM's attempts at enforcing SMP/E dialogs in the late 80s and 'custompak' etc. in the late 90s.) What 'problem' with JCL is fixed by replacing it with REXX? Chris Poncelet CEng MBCS CITP (retired sysprog consultant)
On 12/07/2018 17:38, Nightwatch RenBand wrote: > My information may be out of date... but as I remember someone, possibly > Barry Merrill, did the research and found that running a program under TSO > cost about three times what it cost in batch. And REXX's are usually run > in a TSO environment, even when run as TSO-Batch. > Has this changed? > Perhaps my assumptions about TSO-batch are inaccurate. > But if correct, that pretty much answers why REXX is not a good replacement > for JCL. Yes, JCL is strange and takes some getting used to, but no more > so than English for Chinese speakers, or visa versa. And giving JCL its > due, the overhead it uses is minimal. > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN > . > ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
