On Thu, 13 Jan 2022 21:15:32 +0000, PINION, RICHARD W. <[email protected]> wrote:
>Sequential datasets that they are updated in place cannot be compressed. That >would apply to zEDC or SMS compression. > > Of course not....but the D/R "ready" backup copy (hopefully multiple versions - not the "replicated flavor", once "gold-copy" corrupted, then soon-after all flavors corrupted) can likely be compressed !! Scott Barry SBBTech LLC >-----Original Message----- >From: IBM Mainframe Discussion List <[email protected]> On Behalf Of >Scott Barry >Sent: Thursday, January 13, 2022 4:10 PM >To: [email protected] >Subject: Re: zEDC Justification documents or links > >[External Email. Exercise caution when clicking links or opening attachments.] > >On Thu, 13 Jan 2022 19:02:16 +0000, Richards, Robert B. (CTR) ><[email protected]> wrote: > >>Thanks for the additional information. I'll file it away if we get a z15. On >>a z14_ZR1 at the moment. >> >>-----Original Message----- >>From: IBM Mainframe Discussion List <[email protected]> On >>Behalf Of Scott Barry >>Sent: Thursday, January 13, 2022 1:56 PM >>To: [email protected] >>Subject: Re: zEDC Justification documents or links >> >>On Thu, 13 Jan 2022 16:57:40 +0000, Richards, Robert B. (CTR) >><[email protected]> wrote: >> >>>Anyone have some good links to share that will state why zEDC is worth >>>licensing and getting? 😊 >>> >>>For the record, we already have the HW feature installed. Don’t ask. >>><sigh> >> >>Consider with the IBM z15, there is no longer any "license requirement" for >>activation - no longer implemented with (z14, z13) hardware PCIe cards >>(recommended: at least 4 per CPC/CEC) involved, as the function is on-chip. >>And the performance measurement moves from RMF 74 subtype 9 to subtype 10; >>also, some of the SMF 30 metrics are zeroed-out as no longer relevant. >> >>Scott Barry >>SBBTech LLC >> > >Additional technical information about zEDC has been reported over-time in the >Watson & Walker Tuning Newsletters, such as 2016-4. Also, explore past SHARE >presentations (e.g., circa Anaheim 2014). > >Savings opportunities abound (e.g., SMF compression 8-to-1 within LOGSTREAM >and offloaded to DASD), with mgmt areas like SMF (LOGSTREAMs), z/OS sequential >(DASD-output - look to convert z/OS COMPACTION leverage zEDC, using new SMS >DATACLAS like DCZEDCPR) datasets, DFHSM (data migration/backup-copy), DFDSS >(ADRDSSU - consider ZAP to use ZCOMPRESS activated by-default); also, ISVs >with FDR (config setting ZEDC=YES), PKZIP (also config setting for global >use), IBM/Sterling Connect:Direct, XCOM; also with MQ logging, and beyond >z/OS with JAVA (Z-deployment) and zlib. Also, look to zBNA tool for dataset >candidate identification - not limited to DASD, but also consider tape-to-DASD >migration with zEDC as justification-point. > >You will need/want to gather z/OS performance metrics / reporting setup in >order to keep tabs on zEDC card-performance with utilization, queueing, and >also to consider opportunity to stagger some "time-period demand", such as SMF >LOGSTREAM offloading processes, so as not over-tax PCIe card utilization. > > >Scott Barry >SBBTech LLC > ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
