Thank-you for this feedback. I’m starting to feel like this is a half-baked solution looking for a problem. I know of no way to systematically code for open for update….
Dave Jousma Vice President | Director, Technology Engineering From: IBM Mainframe Discussion List <[email protected]> on behalf of rpinion865 <[email protected]> Date: Tuesday, April 16, 2024 at 12:29 PM To: [email protected] <[email protected]> Subject: Re: Anyone exploiting ZEDC? At a prior life, we got the zEDC cards on a z15, and turned that on for PS datasets. But like you, we quickly learned that PS datasets opened for in/out processing (update in place) did not work. As a compromise, we narrowed down to new PS At a prior life, we got the zEDC cards on a z15, and turned that on for PS datasets. But like you, we quickly learned that PS datasets opened for in/out processing (update in place) did not work. As a compromise, we narrowed down to new PS GDG datasets. Also we looked at new PS GDG datasets over a certain size. We picked an arbitrary number of tracks/cylinders to apply the zEDC compression to. Once we did that, our production job abends disappeared and we started seeing a reduction in space usage on the storage groups where these datasets were being allocated on. Sent with Proton Mail secure email. On Tuesday, April 16th, 2024 at 12:16 PM, Jousma, David <[email protected]> wrote: > Is anyone exploiting ZEDC data compression accelerator in your environments? > We recently licensed the enablement and are working through the issues in our > DEV environment. > > We initially enabled Extended Format/COMPACT ZP, for all DSORG PS datasets, > but are quickly finding that DFSORT, SAS, ISPF recovery datasets all have > issues. We’ve turned back off for now. > > There is no central location in any IBM doc(including recent REDBOOKS) that > discusses where we can and cannot leverage ZEDC. As it stands now, we had to > back out, and looking at taking a new approach in our ACS routines checking > for DSORG=PS, and then if not temp-dataset, or tape assign a DC with the > right options. > > What I am wondering is if anyone is further along that can share how they > rolled out? I really don’t want to have to make the applications teams have > to “opt-in” by coding a DC in their JCL or other dataset allocations. > > Dave Jousma > Vice President | Director, Technology Engineering > > > > > > This e-mail transmission contains information that is confidential and may be > privileged. It is intended only for the addressee(s) named above. If you > receive this e-mail in error, please do not read, copy or disseminate it in > any manner. If you are not the intended recipient, any disclosure, copying, > distribution or use of the contents of this information is prohibited. Please > reply to the message immediately by informing the sender that the message was > misdirected. After replying, please erase it from your computer system. Your > assistance in correcting this error is appreciated. > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN This e-mail transmission contains information that is confidential and may be privileged. It is intended only for the addressee(s) named above. If you receive this e-mail in error, please do not read, copy or disseminate it in any manner. If you are not the intended recipient, any disclosure, copying, distribution or use of the contents of this information is prohibited. Please reply to the message immediately by informing the sender that the message was misdirected. After replying, please erase it from your computer system. Your assistance in correcting this error is appreciated. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
