Short answer - Compression is about 10X more expensive than decompression. Compression is done by examining the first X amount of the data and building a compression dictionary. Once the dictionary is built, the rest of the data is compressed using that data. Compressing data you are going to read a lot makes sense, unless DASD space is a real problem.
Chris Blaicher Technical Architect Software Development Syncsort Incorporated 50 Tice Boulevard, Woodcliff Lake, NJ 07677 P: 201-930-8234 | M: 512-627-3803 E: cblaic...@syncsort.com -----Original Message----- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Massimo Biancucci Sent: Tuesday, May 05, 2015 10:39 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: DFSMS Compression. Hi everybody, occasionally I wonder how SMS Compression (Compaction=YES on DATACLASS) works and how much, in terms of elapsed and CPU, does it modify the step execution. In my shop I usually observe the obvious increase of CPU usage and a (maybe less obvious) bigger elapsed time if compared to tape write. In this "real example" I ran the same program with same input (a SMS compressed dataset) writing to a SMS Compressed dataset, a DUMMY DD and a Virtual Tape dataset. The output dataset has more than 930 millions records. The final compression ratio is more than 70%. Looking at the step statistic: Output - Compressed Dataset STEPNAME PROCSTEP RC EXCP CPU SRB CLOCK SERV PG PAGE SWAP ST040 00 1247K 8.65 .79 23.77 38592K 0 0 0 Output - DUMMY STEPNAME PROCSTEP RC EXCP CPU SRB CLOCK SERV PG PAGE SWAP ST040 00 201K 1.10 .04 2.24 4725K 0 0 0 Output - Virtual Tape Dataset STEPNAME PROCSTEP RC EXCP CPU SRB CLOCK SERV PG PAGE SWAP ST040 00 3730K 1.19 .20 8.86 7476K 0 0 0 What is not so clear to me is why the elapsed time does grow so much even though there's no CPU constraint. Because of the lower number of EXPCs I'd been expecting something less in terms of elapsed or something similar. An A.P.A. trace on the job while writing Compressed Dataset shows that more then 80% of the TCB time in on Compression Services (of course it depends on the fact the application itself uses a small amount of CPU). I wasn't able to find out any further infos about how DFSMS compression works. Finally, sometimes (observing other jobs) the behaviour is not so regular. Thanks in advance for your hints. Massimo ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN ________________________________ ATTENTION: ----- The information contained in this message (including any files transmitted with this message) may contain proprietary, trade secret or other confidential and/or legally privileged information. Any pricing information contained in this message or in any files transmitted with this message is always confidential and cannot be shared with any third parties without prior written approval from Syncsort. This message is intended to be read only by the individual or entity to whom it is addressed or by their designee. If the reader of this message is not the intended recipient, you are on notice that any use, disclosure, copying or distribution of this message, in any form, is strictly prohibited. If you have received this message in error, please immediately notify the sender and/or Syncsort and destroy all copies of this message in your possession, custody or control. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
