It is often quoted mainframe files compress 4 to 1. Might already be compressed, and vsam overhead might not compress as well. So I would try 20%, 10% of input file size, and be sure to include enough free space on enough volumes for the base file size to spread out. PS has a 15 or 16 extent limit across all volumes.
On Fri, Dec 19, 2025 at 12:00 PM Jon Perryman <[email protected]> wrote: > > On Fri, 19 Dec 2025 19:06:45 +0400, Peter <[email protected]> wrote: > > >Is there a sample JCL to dump the dataset to a compressed multi volume > >dataset ? > > Doc for dump logical dataset > https://www.ibm.com/docs/en/zos/2.1.0?topic=dfsmsdss-dump-dataset-syntax-logical-data-set > which should be reviewed for options you may need. You'll find RESTORE in > this same manual. > > //STEP1 EXEC PGM=ADRDSSU > //DUMPDD DD DSN=BACKUP.DSN,DISP=(NEW,CATLG), > // SPACE=(CYL,(50000,50000),RLSE) > //SYSPRINT DD SYSOUT=A > //SYSIN DD * > DUMP OUTDD(DUMPDD) - > DS(INCL(DATASET.TO.DUMP)) - > /* TOL(ENQF) WAIT(0,0) /* if you can't quiesce the dataset and accept the > risks */ - > COMPRESS /* or HWCOMPRESS */ > /* > > >So what would be the recommended size to dump on a compressed dataset? > > Since it's a vsam dataset and you are compressing the backup to use as little > space as possible. Additionally, ADRDSSU uses full track. Look at the options > to see if there is anything relevant to your situation. > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to [email protected] with the message: INFO IBM-MAIN -- Mike A Schwab, Springfield IL USA Where do Forest Rangers go to get away from it all? ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
