Re: Internal Coupling Channel on z13
I disagree. The answer is "it depends". -Original Message- From: IBM Mainframe Discussion List On Behalf Of Seymour J Metz Sent: Tuesday, February 5, 2019 2:50 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: Internal Coupling Channel on z13 It is possible to convert ENQ with SYSTEMS to RESERVE, but the performance ramifications are unacceptable. -- Shmuel (Seymour J.) Metz https://apac01.safelinks.protection.outlook.com/?url=http:%2F%2Fmason.gmu.edu%2F~smetz3data=02%7C01%7Callan.staller%40HCL.COM%7Cd460b1fa54164846cc9608d68bab8303%7C189de737c93a4f5a8b686f4ca9941912%7C0%7C0%7C636849966118974286sdata=qJCwpw%2F9ywQBZLGUH5jHGRaW5FRG5mGzCkE6%2Bc3umkw%3Dreserved=0 From: IBM Mainframe Discussion List on behalf of Tony Harminc Sent: Monday, February 4, 2019 3:42 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: Internal Coupling Channel on z13 On Wed, 30 Jan 2019 at 01:53, Ed Jaffe wrote: > Is it no longer possible to use "old school" shared DASD > RESERVE/RELEASE to protect data? I know it won't work for sharing > PDSE, but for old-school PDS and sequential, it should still work. Reserve/Release works only if someone issues those CCWs. DADSM issues them to protect the VTOC, and maybe catalog management does (I don't know), and the Binder does if SYSLMOD is on a shared device. But data management generally does not, so e.g. if you have two jobs on separate LPARs and each has DISP=OLD for the same dataset using QSAM or BPAM, say, nothing protects the data itself from being written from both sides at the same time. You can put the Reserve/Release in your application program (via ENQ), but that's not going to work very well for your typical COBOL program, and unless you have extreme data separation by volume, performance will suck, to put it politely. Typically what is wanted is the notion of ENQ with the SYSTEMS option, and that's just not available (and is silently ignored if requested) without either GRS or one of the products popular in the 1980s that implemented GRS-like behaviour using Reserve/Release on some kind of control dataset. I don't know if any of those products are still on the market. Or why not, as another poster suggested, use a GRS ring. This can protect applications against dataset corruption at on-the-box CTC speed, and I don't see any obvious reason if would be slower than a timeshared CF on the same machine. Of course you get GRS - not Parallel Sysplex, and there are various things such as JES2 shared SPOOL that need the Sysplex. Tony H. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN ::DISCLAIMER:: -- The contents of this e-mail and any attachment(s) are confidential and intended for the named recipient(s) only. E-mail transmission is not guaranteed to be secure or error-free as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or may contain viruses in transmission. The e mail and its contents (with or without referred errors) shall therefore not attach any liability on the originator or HCL or its affiliates. Views or opinions, if any, presented in this email are solely those of the author and may not necessarily reflect the views or opinions of HCL or its affiliates. Any form of reproduction, dissemination, copying, disclosure, modification, distribution and / or publication of this message without the prior written consent of authorized representative of HCL is strictly prohibited. If you have received this email in error please delete it and notify the sender immediately. Before opening any email and/or attachments, please check them for viruses and other defects. -- -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
It is possible to convert ENQ with SYSTEMS to RESERVE, but the performance ramifications are unacceptable. -- Shmuel (Seymour J.) Metz http://mason.gmu.edu/~smetz3 From: IBM Mainframe Discussion List on behalf of Tony Harminc Sent: Monday, February 4, 2019 3:42 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: Internal Coupling Channel on z13 On Wed, 30 Jan 2019 at 01:53, Ed Jaffe wrote: > Is it no longer possible to use "old school" shared DASD RESERVE/RELEASE > to protect data? I know it won't work for sharing PDSE, but for > old-school PDS and sequential, it should still work. Reserve/Release works only if someone issues those CCWs. DADSM issues them to protect the VTOC, and maybe catalog management does (I don't know), and the Binder does if SYSLMOD is on a shared device. But data management generally does not, so e.g. if you have two jobs on separate LPARs and each has DISP=OLD for the same dataset using QSAM or BPAM, say, nothing protects the data itself from being written from both sides at the same time. You can put the Reserve/Release in your application program (via ENQ), but that's not going to work very well for your typical COBOL program, and unless you have extreme data separation by volume, performance will suck, to put it politely. Typically what is wanted is the notion of ENQ with the SYSTEMS option, and that's just not available (and is silently ignored if requested) without either GRS or one of the products popular in the 1980s that implemented GRS-like behaviour using Reserve/Release on some kind of control dataset. I don't know if any of those products are still on the market. Or why not, as another poster suggested, use a GRS ring. This can protect applications against dataset corruption at on-the-box CTC speed, and I don't see any obvious reason if would be slower than a timeshared CF on the same machine. Of course you get GRS - not Parallel Sysplex, and there are various things such as JES2 shared SPOOL that need the Sysplex. Tony H. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
On Wed, 30 Jan 2019 at 01:10, Brian Westerman wrote: > > Do you have any figures for how much "more" friendly the CPU usage is? Funny thing... More than 40 years ago VM/370 was able to detect the standard TIO/BC loop when issued in a virtual machine, and instead of allowing it to eat CPU time, dispatched other work until the needed I/O interrupt occurred or the virtual device otherwise became available. One might think that PR/SM or whatever that code is called these days could do the analogous thing without needing any changes to the coupling code... Tony H. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
On Wed, 30 Jan 2019 at 01:53, Ed Jaffe wrote: > Is it no longer possible to use "old school" shared DASD RESERVE/RELEASE > to protect data? I know it won't work for sharing PDSE, but for > old-school PDS and sequential, it should still work. Reserve/Release works only if someone issues those CCWs. DADSM issues them to protect the VTOC, and maybe catalog management does (I don't know), and the Binder does if SYSLMOD is on a shared device. But data management generally does not, so e.g. if you have two jobs on separate LPARs and each has DISP=OLD for the same dataset using QSAM or BPAM, say, nothing protects the data itself from being written from both sides at the same time. You can put the Reserve/Release in your application program (via ENQ), but that's not going to work very well for your typical COBOL program, and unless you have extreme data separation by volume, performance will suck, to put it politely. Typically what is wanted is the notion of ENQ with the SYSTEMS option, and that's just not available (and is silently ignored if requested) without either GRS or one of the products popular in the 1980s that implemented GRS-like behaviour using Reserve/Release on some kind of control dataset. I don't know if any of those products are still on the market. Or why not, as another poster suggested, use a GRS ring. This can protect applications against dataset corruption at on-the-box CTC speed, and I don't see any obvious reason if would be slower than a timeshared CF on the same machine. Of course you get GRS - not Parallel Sysplex, and there are various things such as JES2 shared SPOOL that need the Sysplex. Tony H. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
I would form a view as to how important CF request performance is for Dev. If it's important I'd be tempted to turn DYNDISP off for Dev and let Sandbox suffer. Cheers, Martin Martin Packer zChampion, Systems Investigator & Performance Troubleshooter, IBM +44-7802-245-584 email: martin_pac...@uk.ibm.com Twitter / Facebook IDs: MartinPacker Blog: https://www.ibm.com/developerworks/mydeveloperworks/blogs/MartinPacker Podcast Series (With Marna Walle): https://developer.ibm.com/tv/mpt/or https://itunes.apple.com/gb/podcast/mainframe-performance-topics/id1127943573?mt=2 Youtube channel: https://www.youtube.com/channel/UCu_65HaYgksbF6Q8SQ4oOvA From: Jesse 1 Robinson To: IBM-MAIN@LISTSERV.UA.EDU Date: 31/01/2019 18:44 Subject:Re: Internal Coupling Channel on z13 Sent by:IBM Mainframe Discussion List We share CFs between Sandbox and Dev. The latter is not classified here as Prod, though its usage is a lot more prod-ish than is Sandbox. We currently set all the shared-CP LPARs to THIN based on some Q in a SHARE session. Should we revisit that question? . . J.O.Skip Robinson Southern California Edison Company Electric Dragon Team Paddler SHARE MVS Program Co-Manager 323-715-0595 Mobile 626-543-6132 Office ⇐=== NEW robin...@sce.com -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Martin Packer Sent: Thursday, January 31, 2019 1:38 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: (External):Re: Internal Coupling Channel on z13 (This thread has got quite long so pardon me if I repeat something someone else said.) If you must run a PRODUCTION Coupling Facility LPAR on SHARED engines I would generally recommend turning DYNDISP off for that one LPAR. Set it to THIN for the non-Production ones it's forced to share with. The reason is you don't want the Production CF to give up the engine in a particularly timely fashion - because another request might come in soon after the last one. For the Non-Prod ones you want THIN because you want them to get out of the way ASAP. However, the fact of sharing at all is the major destructive factor. THIN just mitigates it. (The instrumentation to guide you on this is SMF74INT, R744PBSY and R744PWAI - which can be examined down to the engine level. Sometimes with value.) But, as always, actual mainframe estate shape governs the decisions. Cheers, Martin Martin Packer zChampion, Systems Investigator & Performance Troubleshooter, IBM +44-7802-245-584 email: martin_pac...@uk.ibm.com Twitter / Facebook IDs: MartinPacker Blog: https://www.ibm.com/developerworks/mydeveloperworks/blogs/MartinPacker Podcast Series (With Marna Walle): https://developer.ibm.com/tv/mpt/or https://urldefense.proofpoint.com/v2/url?u=https-3A__itunes.apple.com_gb_podcast_mainframe-2Dperformance-2Dtopics_id1127943573-3Fmt-3D2=DwIGaQ=jf_iaSHvJObTbx-siA1ZOg=BsPGKdq7-Vl8MW2-WOWZjlZ0NwmcFSpQCLphNznBSDQ=Kc_UcJWF9tIwa5POuK33JinEdDOOQUM6djqCZa7uYTo=72EebpfuqG6c8aHIvYYSvia9yw4fcOi8awkWGDnFY6o= Youtube channel: https://urldefense.proofpoint.com/v2/url?u=https-3A__www.youtube.com_channel_UCu-5F65HaYgksbF6Q8SQ4oOvA=DwIGaQ=jf_iaSHvJObTbx-siA1ZOg=BsPGKdq7-Vl8MW2-WOWZjlZ0NwmcFSpQCLphNznBSDQ=Kc_UcJWF9tIwa5POuK33JinEdDOOQUM6djqCZa7uYTo=NMd8FE00CVO_bO7Ac23QRR1b8oyZSR2vWQwPLwwXG9I= From: Ravi Gaur To: IBM-MAIN@LISTSERV.UA.EDU Date: 31/01/2019 08:46 Subject:Re: Internal Coupling Channel on z13 Sent by:IBM Mainframe Discussion List DYNDISP is recommended with THIN INTERRUPT for Non dedicated CF CPU ...like for our sandbox/dev systems we have non dedicated and we keep it thin for them while for production which has CF CPU dedicated it's good idea to keep it OFF ...You may like to look at Z14 Configuration setup guide which well explained it and obviously to make these changes you will have to do it from HMC . Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
We share CFs between Sandbox and Dev. The latter is not classified here as Prod, though its usage is a lot more prod-ish than is Sandbox. We currently set all the shared-CP LPARs to THIN based on some Q in a SHARE session. Should we revisit that question? . . J.O.Skip Robinson Southern California Edison Company Electric Dragon Team Paddler SHARE MVS Program Co-Manager 323-715-0595 Mobile 626-543-6132 Office ⇐=== NEW robin...@sce.com -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Martin Packer Sent: Thursday, January 31, 2019 1:38 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: (External):Re: Internal Coupling Channel on z13 (This thread has got quite long so pardon me if I repeat something someone else said.) If you must run a PRODUCTION Coupling Facility LPAR on SHARED engines I would generally recommend turning DYNDISP off for that one LPAR. Set it to THIN for the non-Production ones it's forced to share with. The reason is you don't want the Production CF to give up the engine in a particularly timely fashion - because another request might come in soon after the last one. For the Non-Prod ones you want THIN because you want them to get out of the way ASAP. However, the fact of sharing at all is the major destructive factor. THIN just mitigates it. (The instrumentation to guide you on this is SMF74INT, R744PBSY and R744PWAI - which can be examined down to the engine level. Sometimes with value.) But, as always, actual mainframe estate shape governs the decisions. Cheers, Martin Martin Packer zChampion, Systems Investigator & Performance Troubleshooter, IBM +44-7802-245-584 email: martin_pac...@uk.ibm.com Twitter / Facebook IDs: MartinPacker Blog: https://www.ibm.com/developerworks/mydeveloperworks/blogs/MartinPacker Podcast Series (With Marna Walle): https://developer.ibm.com/tv/mpt/or https://itunes.apple.com/gb/podcast/mainframe-performance-topics/id1127943573?mt=2 Youtube channel: https://www.youtube.com/channel/UCu_65HaYgksbF6Q8SQ4oOvA From: Ravi Gaur To: IBM-MAIN@LISTSERV.UA.EDU Date: 31/01/2019 08:46 Subject:Re: Internal Coupling Channel on z13 Sent by:IBM Mainframe Discussion List DYNDISP is recommended with THIN INTERRUPT for Non dedicated CF CPU ...like for our sandbox/dev systems we have non dedicated and we keep it thin for them while for production which has CF CPU dedicated it's good idea to keep it OFF ...You may like to look at Z14 Configuration setup guide which well explained it and obviously to make these changes you will have to do it from HMC . Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
Issuing DYNDISP THIN command on a CF with no shared CF engines gets CF0505I DYNDISP command cancelled . . J.O.Skip Robinson Southern California Edison Company Electric Dragon Team Paddler SHARE MVS Program Co-Manager 323-715-0595 Mobile 626-543-6132 Office ⇐=== NEW robin...@sce.com -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Ravi Gaur Sent: Thursday, January 31, 2019 12:46 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: (External):Re: Internal Coupling Channel on z13 DYNDISP is recommended with THIN INTERRUPT for Non dedicated CF CPU ...like for our sandbox/dev systems we have non dedicated and we keep it thin for them while for production which has CF CPU dedicated it's good idea to keep it OFF ...You may like to look at Z14 Configuration setup guide which well explained it and obviously to make these changes you will have to do it from HMC . -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
(This thread has got quite long so pardon me if I repeat something someone else said.) If you must run a PRODUCTION Coupling Facility LPAR on SHARED engines I would generally recommend turning DYNDISP off for that one LPAR. Set it to THIN for the non-Production ones it's forced to share with. The reason is you don't want the Production CF to give up the engine in a particularly timely fashion - because another request might come in soon after the last one. For the Non-Prod ones you want THIN because you want them to get out of the way ASAP. However, the fact of sharing at all is the major destructive factor. THIN just mitigates it. (The instrumentation to guide you on this is SMF74INT, R744PBSY and R744PWAI - which can be examined down to the engine level. Sometimes with value.) But, as always, actual mainframe estate shape governs the decisions. Cheers, Martin Martin Packer zChampion, Systems Investigator & Performance Troubleshooter, IBM +44-7802-245-584 email: martin_pac...@uk.ibm.com Twitter / Facebook IDs: MartinPacker Blog: https://www.ibm.com/developerworks/mydeveloperworks/blogs/MartinPacker Podcast Series (With Marna Walle): https://developer.ibm.com/tv/mpt/or https://itunes.apple.com/gb/podcast/mainframe-performance-topics/id1127943573?mt=2 Youtube channel: https://www.youtube.com/channel/UCu_65HaYgksbF6Q8SQ4oOvA From: Ravi Gaur To: IBM-MAIN@LISTSERV.UA.EDU Date: 31/01/2019 08:46 Subject:Re: Internal Coupling Channel on z13 Sent by:IBM Mainframe Discussion List DYNDISP is recommended with THIN INTERRUPT for Non dedicated CF CPU ...like for our sandbox/dev systems we have non dedicated and we keep it thin for them while for production which has CF CPU dedicated it's good idea to keep it OFF ...You may like to look at Z14 Configuration setup guide which well explained it and obviously to make these changes you will have to do it from HMC . -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
DYNDISP is recommended with THIN INTERRUPT for Non dedicated CF CPU ...like for our sandbox/dev systems we have non dedicated and we keep it thin for them while for production which has CF CPU dedicated it's good idea to keep it OFF ...You may like to look at Z14 Configuration setup guide which well explained it and obviously to make these changes you will have to do it from HMC . -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
Were this on single CP systems or did you have multiple CP's or specialty processors handing things? Brian -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
On 1/30/2019 3:19 PM, Jesse 1 Robinson wrote: Hence we were running with the default DYNDISP OFF. We set DYNDISP to THIN INTERRUPT via CF command, and suddenly all was well again. Night and day. Haha! No kidding! DYNDISP=OFF is basically a TIGHT CPU LOOP! -- Phoenix Software International Edward E. Jaffe 831 Parkview Drive North El Segundo, CA 90245 https://www.phoenixsoftware.com/ This e-mail message, including any attachments, appended messages and the information contained therein, is for the sole use of the intended recipient(s). If you are not an intended recipient or have otherwise received this email message in error, any use, dissemination, distribution, review, storage or copying of this e-mail message and the information contained therein is strictly prohibited. If you are not an intended recipient, please contact the sender by reply e-mail and destroy all copies of this email message and do not otherwise utilize or retain this email message or any or all of the information contained therein. Although this email message and any attachments or appended messages are believed to be free of any virus or other defect that might affect any computer system into which it is received and opened, it is the responsibility of the recipient to ensure that it is virus free and no responsibility is accepted by the sender for any loss or damage arising in any way from its opening or use. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
(Mentioned this in a previous post.) We replaced two CECs recently. Once everything was running, we discovered that the two-member sandbox was running horribly. Unusable really. Turned out to be configuration. We had copied CEC, LPAR, and Load profiles from old to new boxes. But there is no exportable 'profile' to govern shared CF, i.e. no DYNDISP. Hence we were running with the default DYNDISP OFF. We set DYNDISP to THIN INTERRUPT via CF command, and suddenly all was well again. Night and day. . . J.O.Skip Robinson Southern California Edison Company Electric Dragon Team Paddler SHARE MVS Program Co-Manager 323-715-0595 Mobile 626-543-6132 Office ⇐=== NEW robin...@sce.com -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Ed Jaffe Sent: Tuesday, January 29, 2019 11:29 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: (External):Re: Internal Coupling Channel on z13 On 1/29/2019 10:07 PM, Brian Westerman wrote: > No, just one single CP, no specialty processors are available. We have two CF LPARs (CF01 and CF02) sharing a single ICF engine. From both CF consoles I see: 2019029 22:56:50 => display dyndisp 2019029 22:56:50 CF0512I Dynamic CF Dispatching is THINinterrupts. Right now things are pretty quiet on the system and from my z13s monitoring dashboard, I am seeing our ICF only 1% utilized with THIN interrupts. When I change to DYNDISP ON instead of DYNDISP THIN, I still see only 1% utilized. Of course, when I switch to DYNDISP OFF utilization jumps to 100%. DYNDISP THIN should be more responsive and less CPU hungry than DYNDISP ON, but I can't really tell the difference with my simple test. Both seem pretty thrifty. -- Phoenix Software International Edward E. Jaffe 831 Parkview Drive North El Segundo, CA 90245 https://www.phoenixsoftware.com/ -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
I haven't seen many single-CP boxes in general, and haven't seen one running both CFCC and z/OS on that single CP. My expectation is that this would perform poorly. Sync requests would be impossible since PR/SM can't have both the z/OS and CFCC dispatched on the single CP at the same time, so all requests would have to be converted to async. My expectation is that just using CTCs would probably be faster. I would exercise caution in testing this though, and if it was my production system I probably wouldn't even try. But I'd certainly be curious about the results if you do try it. Note that my reluctance applies specifically to single-CP machines. If I had even two CPs, then I'd be much more willing to give it a shot, depending on current utilization levels and so forth. With dyndisp=thin of course. Scott Chapman On Wed, 30 Jan 2019 00:10:26 -0600, Brian Westerman wrote: >Do you have any figures for how much "more" friendly the CPU usage is? > >This box is a single CPU, no ICF, Zipp or zapp. > >Brian > >-- >For IBM-MAIN subscribe / signoff / archive access instructions, >send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
Others have already said this. With just a single CP on the system, its not a good idea to have a CFCC LPAR on the same CP which is also being used for z/OS LPARs. If you have not already read the the doc at the link below, I suggest you do. It is very comprehensive and covers a lot more than you need. https://www.google.com/url?sa=t=j==s=web=2=2ahUKEwiA3vqJlpXgAhXPc98KHdhGCv8QFjABegQIBxAC=https%3A%2F%2Fwww.ibm.com%2Fdownloads%2Fcas%2FJZB2E38Q=AOvVaw3YAB90hstZWbpesUsL-IDn If the link doesn't work, search for ZSW01971-USEN-26 on google. In addition to this doc, there are various Redbooks and of course Knowledge Center:-) I know you are looking for possible use cases/recommendations for what you would like to do with a single CP system. I will be very surprised if anyone has such a configuration on a 'production' system. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
To: " If they can't afford a second CP, they can't afford an ICF." AFAIK an ICF is cheaper than a CP (I believe half the price). Met vriendelijke groet, Kees Vernooij KLM Information Services z/OS Systems Tel +31 6 10 14 58 78 > -Original Message- > From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On > Behalf Of Ed Jaffe > Sent: 30 January, 2019 7:53 > To: IBM-MAIN@LISTSERV.UA.EDU > Subject: Re: Internal Coupling Channel on z13 > > On 1/29/2019 10:17 PM, Brian Westerman wrote: > > This particular box has just a single CP, no specialty processors, 3 > LPARs, one of them production, one application programmer test, and the > other a sandbox that is extremely low use and in any case shares only > the res volume. > > > > They "need" to run GRS because it's not really safe to run without it > (especially since there is nothing to keep a lockout from occurring but > to be careful), but they can't afford to add a specialty processor, at > least not at this time. > > > > I'm trying to find out if they can install the microcode CF and > assuming that the CPU use for the CF is now "low" if they could use it > for DASD sharing only. There is no real data sharing involved (no DB2, > no CICS). So we are just talking about GRS shipping the ENQs around. > > In theory, coupling facility thin interrupts is faster than previous CF > technologies on shared CPs: > https://www-03.ibm.com/support/techdocs/atsmastr.nsf/WebIndex/WP102400 > > ICF specialty engines are ridiculously priced (i.e., wy too high). > If they can't afford a second CP, they can't afford an ICF. Too bad, > because that would make all the difference. > > Is it no longer possible to use "old school" shared DASD RESERVE/RELEASE > to protect data? I know it won't work for sharing PDSE, but for > old-school PDS and sequential, it should still work. > > > -- > Phoenix Software International > Edward E. Jaffe > 831 Parkview Drive North > El Segundo, CA 90245 > https://www.phoenixsoftware.com/ > > > > > This e-mail message, including any attachments, appended messages and > the > information contained therein, is for the sole use of the intended > recipient(s). If you are not an intended recipient or have otherwise > received this email message in error, any use, dissemination, > distribution, > review, storage or copying of this e-mail message and the information > contained therein is strictly prohibited. If you are not an intended > recipient, please contact the sender by reply e-mail and destroy all > copies > of this email message and do not otherwise utilize or retain this email > message or any or all of the information contained therein. Although > this > email message and any attachments or appended messages are believed to > be > free of any virus or other defect that might affect any computer system > into > which it is received and opened, it is the responsibility of the > recipient > to ensure that it is virus free and no responsibility is accepted by the > sender for any loss or damage arising in any way from its opening or > use. > > -- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN For information, services and offers, please visit our web site: http://www.klm.com. This e-mail and any attachment may contain confidential and privileged material intended for the addressee only. If you are not the addressee, you are notified that no part of the e-mail or any attachment may be disclosed, copied or distributed, and that any other action related to this e-mail or attachment is strictly prohibited, and may be unlawful. If you have received this e-mail by error, please notify the sender immediately by return e-mail, and delete this message. Koninklijke Luchtvaart Maatschappij NV (KLM), its subsidiaries and/or its employees shall not be liable for the incorrect or incomplete transmission of this e-mail or any attachments, nor responsible for any delay in receipt. Koninklijke Luchtvaart Maatschappij N.V. (also known as KLM Royal Dutch Airlines) is registered in Amstelveen, The Netherlands, with registered number 33014286 -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
On 1/29/2019 10:07 PM, Brian Westerman wrote: No, just one single CP, no specialty processors are available. We have two CF LPARs (CF01 and CF02) sharing a single ICF engine. From both CF consoles I see: 2019029 22:56:50 => display dyndisp 2019029 22:56:50 CF0512I Dynamic CF Dispatching is THINinterrupts. Right now things are pretty quiet on the system and from my z13s monitoring dashboard, I am seeing our ICF only 1% utilized with THIN interrupts. When I change to DYNDISP ON instead of DYNDISP THIN, I still see only 1% utilized. Of course, when I switch to DYNDISP OFF utilization jumps to 100%. DYNDISP THIN should be more responsive and less CPU hungry than DYNDISP ON, but I can't really tell the difference with my simple test. Both seem pretty thrifty. -- Phoenix Software International Edward E. Jaffe 831 Parkview Drive North El Segundo, CA 90245 https://www.phoenixsoftware.com/ This e-mail message, including any attachments, appended messages and the information contained therein, is for the sole use of the intended recipient(s). If you are not an intended recipient or have otherwise received this email message in error, any use, dissemination, distribution, review, storage or copying of this e-mail message and the information contained therein is strictly prohibited. If you are not an intended recipient, please contact the sender by reply e-mail and destroy all copies of this email message and do not otherwise utilize or retain this email message or any or all of the information contained therein. Although this email message and any attachments or appended messages are believed to be free of any virus or other defect that might affect any computer system into which it is received and opened, it is the responsibility of the recipient to ensure that it is virus free and no responsibility is accepted by the sender for any loss or damage arising in any way from its opening or use. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
What is the percent busy at peak times? How big a percent do you need? 10% for the ICF partition? Would you save that much by converting Ring to Star? On Wed, Jan 30, 2019 at 12:07 AM Brian Westerman wrote: > > No, just one single CP, no specialty processors are available. > > Brian > > -- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu 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 lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
On 1/29/2019 10:17 PM, Brian Westerman wrote: This particular box has just a single CP, no specialty processors, 3 LPARs, one of them production, one application programmer test, and the other a sandbox that is extremely low use and in any case shares only the res volume. They "need" to run GRS because it's not really safe to run without it (especially since there is nothing to keep a lockout from occurring but to be careful), but they can't afford to add a specialty processor, at least not at this time. I'm trying to find out if they can install the microcode CF and assuming that the CPU use for the CF is now "low" if they could use it for DASD sharing only. There is no real data sharing involved (no DB2, no CICS). So we are just talking about GRS shipping the ENQs around. In theory, coupling facility thin interrupts is faster than previous CF technologies on shared CPs: https://www-03.ibm.com/support/techdocs/atsmastr.nsf/WebIndex/WP102400 ICF specialty engines are ridiculously priced (i.e., wy too high). If they can't afford a second CP, they can't afford an ICF. Too bad, because that would make all the difference. Is it no longer possible to use "old school" shared DASD RESERVE/RELEASE to protect data? I know it won't work for sharing PDSE, but for old-school PDS and sequential, it should still work. -- Phoenix Software International Edward E. Jaffe 831 Parkview Drive North El Segundo, CA 90245 https://www.phoenixsoftware.com/ This e-mail message, including any attachments, appended messages and the information contained therein, is for the sole use of the intended recipient(s). If you are not an intended recipient or have otherwise received this email message in error, any use, dissemination, distribution, review, storage or copying of this e-mail message and the information contained therein is strictly prohibited. If you are not an intended recipient, please contact the sender by reply e-mail and destroy all copies of this email message and do not otherwise utilize or retain this email message or any or all of the information contained therein. Although this email message and any attachments or appended messages are believed to be free of any virus or other defect that might affect any computer system into which it is received and opened, it is the responsibility of the recipient to ensure that it is virus free and no responsibility is accepted by the sender for any loss or damage arising in any way from its opening or use. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
This particular box has just a single CP, no specialty processors, 3 LPARs, one of them production, one application programmer test, and the other a sandbox that is extremely low use and in any case shares only the res volume. They "need" to run GRS because it's not really safe to run without it (especially since there is nothing to keep a lockout from occurring but to be careful), but they can't afford to add a specialty processor, at least not at this time. I'm trying to find out if they can install the microcode CF and assuming that the CPU use for the CF is now "low" if they could use it for DASD sharing only. There is no real data sharing involved (no DB2, no CICS). So we are just talking about GRS shipping the ENQs around. Has anyone tried this yet? Brian -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
I agree that before the current CF20 implementation of the micocode only version, it was supposed ot be almost fatal to try it, but with the new z13s and z14 it's "supposed" to be "low" impact, but it doesn't really talk about how low the impact is, and if it can be done with a single CP and no specialty processors. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
Do you have any figures for how much "more" friendly the CPU usage is? This box is a single CPU, no ICF, Zipp or zapp. Brian -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
NO, I'm trying to use the Microcode Implementation, it has no CTC, and is just memory to memory between LPARs on the same physical box. Brian -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
No, just one single CP, no specialty processors are available. Brian -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
I agree with the recommendation to get a CF engine as the best/first choice all around. I also agree to proceed with caution if you're going to share one or more CPs between the Coupling Facility Control Code (CFCC) and z/OS and/or other operating systems. "Proceed with caution" is not the same thing as "don't." You just want to be extra careful, test well, and back off the idea if it's not suitable in your environment since there are some potential issues that could surface. However, if you have one or more CPs (general purpose processors) that you're willing to *dedicate* to a CFCC LPAR, this caution doesn't apply. In this case you're using a dedicated CP as if it were a CF engine, and that's perfectly fine. You might be dedicating a sub-capacity CP (or more than one) with different capacity characteristics than a CF engine, but there's no particular issue with that as long as you have sufficient capacity for your needs. Or, if you have some "trivial" z/OS workload sharing that CP with the CFCC, such as a "sandbox" z/OS LPAR for system programmers that sees little activity, that might be fine. There are certain upgrade scenarios when it can make a great deal of sense to use one or more CPs for the CFCC for relatively brief periods of time. Timothy Sipples IT Architect Executive, Industry Solutions, IBM Z & LinuxONE E-Mail: sipp...@sg.ibm.com -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
We've used ICF on pretty much every model of hardware since z10. We're currently on z14 and z13s. We use internal coupling links with external links to another CEC. We do have--and always have--CF engines shared among non-prod LPARs using thin interrupt protocol. I would hesitate to suggest buying CFs just for GRS star, but if you already have them, star so far outperforms ring that there should be no hesitation in moving there. I also would not consider using regular CPs. They would work, but poorly. . . J.O.Skip Robinson Southern California Edison Company Electric Dragon Team Paddler SHARE MVS Program Co-Manager 323-715-0595 Mobile 626-543-6132 Office ⇐=== NEW robin...@sce.com -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Dana Mitchell Sent: Tuesday, January 29, 2019 9:28 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: (External):Re: Internal Coupling Channel on z13 Brian, We've actually been thinking about this too, as we evaluate moving to z14's. We currently have a 5 member GRS ring that could really benefit moving to GRS Star, but nobody want's to spend any $ for an ICF. As of BC12 generation IBM was still recommending against using GPs for CF processors: According to an IBM paper titled: Coupling Thin Interrupts and CF Performance in Shared Processor Environments While it is possible to use general purpose processors, CPs, as CF processors and share the CPs with z/OS and other similarly defined CFs, this is not a recommended configuration. Using CPs as CF processors creates interaction between z/OS and the CF in terms of cache reuse and other factors that may impact performance. Dana On Tue, 29 Jan 2019 00:23:56 -0600, Brian Westerman wrote: >Hi, > >Has anyone had any experience with using the internal coupling channels on a >z13. "supposedly" IBM has removed the active wait problems (where the CF lpar >would try to use 100% of whatever it gets from PR/SM), but I was wondering if >it's ready for prime time yet. I have a z13s (single CPU) with 3 LPARs on it, >they are fairly low use, but I think it'a about time the started to use GRS >since they share DASD (it was small, but now more and more as time goes on). >They have been lucky so far, but I would rather not rely on luck. They don't >have the cash to add a specialty processor, and if IBM has really reduced the >overhead of ICF, then it might be a good fit for them. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
Brian, We've actually been thinking about this too, as we evaluate moving to z14's. We currently have a 5 member GRS ring that could really benefit moving to GRS Star, but nobody want's to spend any $ for an ICF. As of BC12 generation IBM was still recommending against using GPs for CF processors: According to an IBM paper titled: Coupling Thin Interrupts and CF Performance in Shared Processor Environments While it is possible to use general purpose processors, CPs, as CF processors and share the CPs with z/OS and other similarly defined CFs, this is not a recommended configuration. Using CPs as CF processors creates interaction between z/OS and the CF in terms of cache reuse and other factors that may impact performance. Dana On Tue, 29 Jan 2019 00:23:56 -0600, Brian Westerman wrote: >Hi, > >Has anyone had any experience with using the internal coupling channels on a >z13. "supposedly" IBM has removed the active wait problems (where the CF lpar >would try to use 100% of whatever it gets from PR/SM), but I was wondering if >it's ready for prime time yet. I have a z13s (single CPU) with 3 LPARs on it, >they are fairly low use, but I think it'a about time the started to use GRS >since they share DASD (it was small, but now more and more as time goes on). >They have been lucky so far, but I would rather not rely on luck. They don't >have the cash to add a specialty processor, and if IBM has really reduced the >overhead of ICF, then it might be a good fit for them. > -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13 [EXTERNAL]
If the lpars are in a base sysplex you can use XCF to carry the GRS traffic and still be in a RING setup. GRS using XCF is better than letting GRS try to handle things. We have been using both an internal and external CF for years. Currently the internal is on a z13. All the lpars on that CEC talk to the CF over internal links. Just for completeness our external CF is sitting on a z13s. Thanks.. Paul Feller AGT Mainframe Technical Support -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Allan Staller Sent: Tuesday, January 29, 2019 7:40 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: Internal Coupling Channel on z13 [EXTERNAL] GRS RING can/will run over CTC's. Not sure I would want to do that in a 3 member ring. It was bad enough in a 2 member ring. I am currently using an ICF on a z/12 BC w/no issue. Howerr, I am not sure an ICF is the same thing you are describing. -Original Message- From: IBM Mainframe Discussion List On Behalf Of Brian Westerman Sent: Tuesday, January 29, 2019 12:24 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Internal Coupling Channel on z13 Hi, Has anyone had any experience with using the internal coupling channels on a z13. "supposedly" IBM has removed the active wait problems (where the CF lpar would try to use 100% of whatever it gets from PR/SM), but I was wondering if it's ready for prime time yet. I have a z13s (single CPU) with 3 LPARs on it, they are fairly low use, but I think it'a about time the started to use GRS since they share DASD (it was small, but now more and more as time goes on). They have been lucky so far, but I would rather not rely on luck. They don't have the cash to add a specialty processor, and if IBM has really reduced the overhead of ICF, then it might be a good fit for them. Anyone using the microcode ICF on a z13 or z14? If so, does it play well with everyone else? Brian -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN ::DISCLAIMER:: -- The contents of this e-mail and any attachment(s) are confidential and intended for the named recipient(s) only. E-mail transmission is not guaranteed to be secure or error-free as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or may contain viruses in transmission. The e mail and its contents (with or without referred errors) shall therefore not attach any liability on the originator or HCL or its affiliates. Views or opinions, if any, presented in this email are solely those of the author and may not necessarily reflect the views or opinions of HCL or its affiliates. Any form of reproduction, dissemination, copying, disclosure, modification, distribution and / or publication of this message without the prior written consent of authorized representative of HCL is strictly prohibited. If you have received this email in error please delete it and notify the sender immediately. Before opening any email and/or attachments, please check them for viruses and other defects. -- -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- Please note: This message originated outside your organization. Please use caution when opening links or attachments. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
W dniu 2019-01-29 o 07:23, Brian Westerman pisze: Hi, Has anyone had any experience with using the internal coupling channels on a z13. "supposedly" IBM has removed the active wait problems (where the CF lpar would try to use 100% of whatever it gets from PR/SM), but I was wondering if it's ready for prime time yet. I have a z13s (single CPU) with 3 LPARs on it, they are fairly low use, but I think it'a about time the started to use GRS since they share DASD (it was small, but now more and more as time goes on). They have been lucky so far, but I would rather not rely on luck. They don't have the cash to add a specialty processor, and if IBM has really reduced the overhead of ICF, then it might be a good fit for them. Anyone using the microcode ICF on a z13 or z14? If so, does it play well with everyone else? Few remarks: 1. Internal channels (ICP) have *nothing to do* with CPU consumption of internal CF. 2. Internal CF is just CF. There is no difference between them, except failure domain. 3. Yes, IBM changed way how CFCC use processor. Instead of legacy "active wait" (means 100% CPU all the time, even not serving anything) there more CPU friendly mode. -- Radoslaw Skorupka Lodz, Poland == Jeśli nie jesteś adresatem tej wiadomości: - powiadom nas o tym w mailu zwrotnym (dziękujemy!), - usuń trwale tę wiadomość (i wszystkie kopie, które wydrukowałeś lub zapisałeś na dysku). Wiadomość ta może zawierać chronione prawem informacje, które może wykorzystać tylko adresat.Przypominamy, że każdy, kto rozpowszechnia (kopiuje, rozprowadza) tę wiadomość lub podejmuje podobne działania, narusza prawo i może podlegać karze. mBank S.A. z siedzibą w Warszawie, ul. Senatorska 18, 00-950 Warszawa,www.mBank.pl, e-mail: kont...@mbank.pl. Sąd Rejonowy dla m. st. Warszawy XII Wydział Gospodarczy Krajowego Rejestru Sądowego, KRS 025237, NIP: 526-021-50-88. Kapitał zakładowy (opłacony w całości) według stanu na 01.01.2018 r. wynosi 169.248.488 złotych. If you are not the addressee of this message: - let us know by replying to this e-mail (thank you!), - delete this message permanently (including all the copies which you have printed out or saved). This message may contain legally protected information, which may be used exclusively by the addressee.Please be reminded that anyone who disseminates (copies, distributes) this message or takes any similar action, violates the law and may be penalised. mBank S.A. with its registered office in Warsaw, ul. Senatorska 18, 00-950 Warszawa,www.mBank.pl, e-mail: kont...@mbank.pl. District Court for the Capital City of Warsaw, 12th Commercial Division of the National Court Register, KRS 025237, NIP: 526-021-50-88. Fully paid-up share capital amounting to PLN 169,248,488 as at 1 January 2018. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
GRS RING can/will run over CTC's. Not sure I would want to do that in a 3 member ring. It was bad enough in a 2 member ring. I am currently using an ICF on a z/12 BC w/no issue. Howerr, I am not sure an ICF is the same thing you are describing. -Original Message- From: IBM Mainframe Discussion List On Behalf Of Brian Westerman Sent: Tuesday, January 29, 2019 12:24 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Internal Coupling Channel on z13 Hi, Has anyone had any experience with using the internal coupling channels on a z13. "supposedly" IBM has removed the active wait problems (where the CF lpar would try to use 100% of whatever it gets from PR/SM), but I was wondering if it's ready for prime time yet. I have a z13s (single CPU) with 3 LPARs on it, they are fairly low use, but I think it'a about time the started to use GRS since they share DASD (it was small, but now more and more as time goes on). They have been lucky so far, but I would rather not rely on luck. They don't have the cash to add a specialty processor, and if IBM has really reduced the overhead of ICF, then it might be a good fit for them. Anyone using the microcode ICF on a z13 or z14? If so, does it play well with everyone else? Brian -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN ::DISCLAIMER:: -- The contents of this e-mail and any attachment(s) are confidential and intended for the named recipient(s) only. E-mail transmission is not guaranteed to be secure or error-free as information could be intercepted, corrupted, lost, destroyed, arrive late or incomplete, or may contain viruses in transmission. The e mail and its contents (with or without referred errors) shall therefore not attach any liability on the originator or HCL or its affiliates. Views or opinions, if any, presented in this email are solely those of the author and may not necessarily reflect the views or opinions of HCL or its affiliates. Any form of reproduction, dissemination, copying, disclosure, modification, distribution and / or publication of this message without the prior written consent of authorized representative of HCL is strictly prohibited. If you have received this email in error please delete it and notify the sender immediately. Before opening any email and/or attachments, please check them for viruses and other defects. -- -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Internal Coupling Channel on z13
Hi Brian. Are you referring to Coupling Facility Thin Interrupts? It sounds like you might be. Also, are you running a dedicated ICF engine? Or sharing? Thanks, Martin Sent from my iPad > On 29 Jan 2019, at 06:24, Brian Westerman wrote: > > Hi, > > Has anyone had any experience with using the internal coupling channels on a z13. "supposedly" IBM has removed the active wait problems (where the CF lpar would try to use 100% of whatever it gets from PR/SM), but I was wondering if it's ready for prime time yet. I have a z13s (single CPU) with 3 LPARs on it, they are fairly low use, but I think it'a about time the started to use GRS since they share DASD (it was small, but now more and more as time goes on). They have been lucky so far, but I would rather not rely on luck. They don't have the cash to add a specialty processor, and if IBM has really reduced the overhead of ICF, then it might be a good fit for them. > > Anyone using the microcode ICF on a z13 or z14? If so, does it play well with everyone else? > > Brian > > -- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN >Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Internal Coupling Channel on z13
Hi, Has anyone had any experience with using the internal coupling channels on a z13. "supposedly" IBM has removed the active wait problems (where the CF lpar would try to use 100% of whatever it gets from PR/SM), but I was wondering if it's ready for prime time yet. I have a z13s (single CPU) with 3 LPARs on it, they are fairly low use, but I think it'a about time the started to use GRS since they share DASD (it was small, but now more and more as time goes on). They have been lucky so far, but I would rather not rely on luck. They don't have the cash to add a specialty processor, and if IBM has really reduced the overhead of ICF, then it might be a good fit for them. Anyone using the microcode ICF on a z13 or z14? If so, does it play well with everyone else? Brian -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN