On Wed, 14 Sep 2011 08:18:12 -0500, McKown, John wrote: >Given that the most obvious difference is that what used to take 3 output >tapes is now going to 12 output tapes, the "slow down" is due to the number of >bytes being written to tape. The bytes read from disk seem be about the same >because the number of cylinders taken by the VSAM dataset is the same. The >"problem" is that Faver is expanding the compressed bytes in the SMS case but >not in the Data Accelerator case. We are talking to CA about why Faver expands >the compressed bytes when the dataset is SMS compressed instead of just >"sucking it up" using something like a "read track" CCW which I think would >bypass the SMS decompression. > >BTW - Faver is working exactly as it is documented to work in the Faver >manual. We just don't like it. <shrug> >
There are 2 new export parameters in the current (latest) Faver version 4.5.0: "SAVECOMP" and "OPTIMIZE". >From the manual: "The SAVECOMP parameter directs CA FAVER to export compressed format KSDS files in their native compressed format. " "The OPTIMIZE parameter is used in conjunction with the SAVECOMP parameter to control the number of tracks read at one time by DFSMSdss when processing hardware compressed files. OPTIMIZE is valid only if SAVECOMP is also specified. If specified, OPTIMIZE must immediately follow the SAVECOMP parameter. " Norbert Friemel ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
