Re: IPL an LPAR with a very low weight?
The customer wants the small LPAR to come up on request only and we want to give the MIPS from the small LPAR to the one weighted at 47, i.e. raise it to 56. It sounds like the best/safest thing to do is NOT make the weight=1. The box usually is not 100% busy, but can be depending on activity. Dave Thorn * Senior Technology Analyst * SunGard Computer Services * 600 Laurel Oak Road, Voorhees, NJ, 08043 Tel 856 566-5412 * Mobile 609 781-0353 * Fax 856 566-3656 CONFIDENTIALITY: This e-mail (including any attachments) may contain confidential, proprietary and privileged information, and unauthorized disclosure or use is prohibited. If you received this e-mail in error, please notify the sender and delete this e-mail from your system. -Original Message- From: Al Sherkow [mailto:[EMAIL PROTECTED] Sent: Friday, October 26, 2007 10:02 PM To: IBM-MAIN@BAMA.UA.EDU; Thorn, David Subject: Re: IPL an LPAR with a very low weight? Hi Dave -- I suspect you wouldn't be considering this unless you were having problems with CPU resource contention in the two big LPARs (at 47 and 43). What kind of machine is this? how many engines? If this small LPAR is down, and the box is nearly 100% busy, IPLing this LPAR will be very slow, and this small LPAR may slow down the others in the sysplex and MIM. What is the real problem you are trying to solve? Al Al Sherkow, I/S Management Strategies, Ltd. Consulting Expertise on Capacity Planning, Performance Tuning, WLC, LPARs, IRD and LCS Software Seminars on IBM SW Pricing, LPARs, and IRD Voice: +1 414 332-3062 Web: www.sherkow.com -- 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
Re: IPL an LPAR with a very low weight?
Our sysprog sandplex has very low weight, like 5% of 1 processor or some such. When the box is full (running 98-100%) not only does it take forever to shut that lpar down, it also takes forever to IPL it. The duration in itself is not the problem, but automation (SA/390 V3) has almost everything in stuck status because things don't terminate in automations' (usually sufficient) timeout-interval. At which point you do a manual shutdown! Coming back up usually isn't as bad, as automation has everything in started2, but recognizes when things are finally 'up'. And don't get me started on XCF communication! We have a permanent vote to keep the automation manager from starting on one system in the sysplex, simply because that one has a lower weight than the others in the plex. Once the automation manager is on that system, XCF communication regularly times out, and then the shutdown is stuck. (This is not the sysprog sandplex!) Regards, Barbara Nitz -- Pt! Schon vom neuen GMX MultiMessenger gehört? Der kann`s mit allen: http://www.gmx.net/de/go/multimessenger -- 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
Re: IPL an LPAR with a very low weight?
Dave Thorn wrote: If an LPAR (in a down state) were set to a very low weight (1, for instance) and then IPLed, would there be problems? This assumes that the other LPARs are not at high utilizations and using all the CPU cycles themselves. Has anyone done this? Have problems occurred? I've been doing this many times. A lot of times. It was always separate system (no GRS complex, no MIM, no shared datasets, almost no shared volumes). I have *never* had any problem with that. In my case 1 means 1%, I tried even less. With or without capping. On litghtly or heavily loaded CPC. on small and big machines (100-1700 MIPS). -- Radoslaw Skorupka Lodz, Poland -- BRE Bank SA ul. Senatorska 18 00-950 Warszawa www.brebank.pl Sd Rejonowy dla m. st. Warszawy XII Wydzia Gospodarczy Krajowego Rejestru Sdowego, nr rejestru przedsibiorców KRS 025237 NIP: 526-021-50-88 Wedug stanu na dzie 01.01.2007 r. kapita zakadowy BRE Banku SA (w caoci opacony) wynosi 118.064.140 z. W zwizku z realizacj warunkowego podwyszenia kapitau zakadowego, na podstawie uchwa XVI WZ z dnia 21.05.2003 r., kapita zakadowy BRE Banku SA moe ulec podwyszeniu do kwoty 118.760.528 z. Akcje w podwyszonym kapitale zakadowym bd w caoci opacone. -- 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
IPL an LPAR with a very low weight?
If an LPAR (in a down state) were set to a very low weight (1, for instance) and then IPLed, would there be problems? This assumes that the other LPARs are not at high utilizations and using all the CPU cycles themselves. Has anyone done this? Have problems occurred? Dave Thorn * Senior Technology Analyst * SunGard Computer Services * 600 Laurel Oak Road, Voorhees, NJ, 08043 Tel 856 566-5412 * Mobile 609 781-0353 * Fax 856 566-3656 CONFIDENTIALITY: This e-mail (including any attachments) may contain confidential, proprietary and privileged information, and unauthorized disclosure or use is prohibited. If you received this e-mail in error, please notify the sender and delete this e-mail from your system. -- 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
Re: IPL an LPAR with a very low weight?
On Fri, 26 Oct 2007 14:44:43 -0400, Dave Thorn wrote: If an LPAR (in a down state) were set to a very low weight (1, for instance) and then IPLed, would there be problems? This assumes that the other LPARs are not at high utilizations and using all the CPU cycles themselves. Has anyone done this? Have problems occurred? Depends, of course. For starters the weight is relative to the aggregate for the CEC - if that is only, say, 2 (with 2 LPARs weighted at 1 each) you ought not expect to have any problems at all. Probably not your case though. Is your LPAR sharing any resources with any others? GRS or MIM come to mind here... if you need to run those on your resource challenged LPAR then you are a troublemaker. If your whimpy LPAR is a monplex, sharing nothing then no, it isn't a problem per se. More of an opportunity to catch up on your reading while you wait for it to respond. (We do it with a sandbox system and it is pretty pokey when the other LPARs are guzzling MIPS. Still, it all works but at an HO-scale. Just not the way to run a real railroad.) -- Tom Schmidt Madison, WI -- 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
Re: IPL an LPAR with a very low weight?
If an LPAR (in a down state) were set to a very low weight (1, for instance) and then IPLed, would there be problems? This assumes that the other LPARs are not at high utilizations and using all the CPU cycles themselves. Weights don't matter when the other LPARs are not at high. They only matter when there is processor contention. When that happens, a 1 may be a problem, depending on what it means relative to the other LPARs. If they add up to 9 (or 3, or whatever), then it should not be an issue. But, if they add up to 999 (or 900, or whatever) then you will likely have a problem. I have found, empirically, that you need about 10% of a processor to get through an IPL. But, that was on slower processors. - Too busy driving to stop for gas! -- 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
Re: IPL an LPAR with a very low weight?
-Original Message- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of Dave Thorn Sent: Friday, October 26, 2007 1:45 PM To: IBM-MAIN@BAMA.UA.EDU Subject: IPL an LPAR with a very low weight? If an LPAR (in a down state) were set to a very low weight (1, for instance) and then IPLed, would there be problems? This assumes that the other LPARs are not at high utilizations and using all the CPU cycles themselves. Has anyone done this? Have problems occurred? I have two LPARs. The production LPAR has a weight of 244. The sysprog LPAR has a weight of 10. I've never had any problem, other than lack of CPU at times in the sysprog LPAR. In the past, we had three LPARs. The third LPAR did some DASD backups and had a small weight. This did cause problems due to enqueue (DASD reserve) contention. This was not due to a small weight, per se. -- John McKown Senior Systems Programmer HealthMarkets Keeping the Promise of Affordable Coverage Administrative Services Group Information Technology The information contained in this e-mail message may be privileged and/or confidential. It is for intended addressee(s) only. If you are not the intended recipient, you are hereby notified that any disclosure, reproduction, distribution or other use of this communication is strictly prohibited and could, in certain circumstances, be a criminal offense. If you have received this e-mail in error, please notify the sender by reply and delete this message without copying or disclosing it. -- 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
Re: IPL an LPAR with a very low weight?
2 other LPARs on this box: one weight is 47 the other is 43. We want to make the 3rd LPAR=1 (down from 10) and give the other 9 to the one at 47. This LPAR shares MIM with one LPAR on this box and 2 other LPARs on 2 other boxes; all are in the same sysplex. So if it IPLs slowly that's where the problems occur? Dave Thorn * Senior Technology Analyst * SunGard Computer Services * 600 Laurel Oak Road, Voorhees, NJ, 08043 Tel 856 566-5412 * Mobile 609 781-0353 * Fax 856 566-3656 CONFIDENTIALITY: This e-mail (including any attachments) may contain confidential, proprietary and privileged information, and unauthorized disclosure or use is prohibited. If you received this e-mail in error, please notify the sender and delete this e-mail from your system. -Original Message- From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf Of Tom Schmidt Sent: Friday, October 26, 2007 2:52 PM To: IBM-MAIN@BAMA.UA.EDU Subject: Re: IPL an LPAR with a very low weight? On Fri, 26 Oct 2007 14:44:43 -0400, Dave Thorn wrote: If an LPAR (in a down state) were set to a very low weight (1, for instance) and then IPLed, would there be problems? This assumes that the other LPARs are not at high utilizations and using all the CPU cycles themselves. Has anyone done this? Have problems occurred? Depends, of course. For starters the weight is relative to the aggregate for the CEC - if that is only, say, 2 (with 2 LPARs weighted at 1 each) you ought not expect to have any problems at all. Probably not your case though. Is your LPAR sharing any resources with any others? GRS or MIM come to mind here... if you need to run those on your resource challenged LPAR then you are a troublemaker. If your whimpy LPAR is a monplex, sharing nothing then no, it isn't a problem per se. More of an opportunity to catch up on your reading while you wait for it to respond. (We do it with a sandbox system and it is pretty pokey when the other LPARs are guzzling MIPS. Still, it all works but at an HO-scale. Just not the way to run a real railroad.) -- Tom Schmidt Madison, WI -- 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 -- 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
Re: IPL an LPAR with a very low weight?
Assuming you are not using a hard (or soft) cap and the other LPARs are not using the whole box, you should be able to run at whatever the others do not use. Because you are using MIM you might want to leave it at 10 and assuming no LPAR is capped and the CEC is not running flat out the other LPARs can use whatever portion of the 10 is not needed. If you do approach 100% you want MIM on the little guy to be reasonably responsive. There are ways to tune MIM for LPARs of different responsiveness (IIRC you end up forcing the little guy to hit the control files more often than demand would normally dictate). -Original Message- Dave Thorn 2 other LPARs on this box: one weight is 47 the other is 43. We want to make the 3rd LPAR=1 (down from 10) and give the other 9 to the one at 47. This LPAR shares MIM with one LPAR on this box and 2 other LPARs on 2 other boxes; all are in the same sysplex. So if it IPLs slowly that's where the problems occur? -- 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
Re: IPL an LPAR with a very low weight?
On Fri, 26 Oct 2007 15:13:41 -0400, Dave Thorn [EMAIL PROTECTED] wrote: 2 other LPARs on this box: one weight is 47 the other is 43. We want to make the 3rd LPAR=1 (down from 10) and give the other 9 to the one at 47. This LPAR shares MIM with one LPAR on this box and 2 other LPARs on 2 other boxes; all are in the same sysplex. So if it IPLs slowly that's where the problems occur? Is it MIA or MII (or both). It doesn't matter that much... but tape allocation delays are not as bad as ENQ delays. Either way, since you are sharing and you can't get MIM higher than SYSSTC (or SYSTEM if using the address space create method of starting it) - it still has to share with all the OS system tasks at the same priority and if the box is maxed out you'll affect your other environments. I would look at RMF reports and see what the idle usage is and certainly not make the weight lower than that. But you would probably want to either bump up the weight during IPLs or start MIM after the rest of the IPL was done. If this was a totally stand alone LPAR, then I agree with what one of the previous posters said... it will just be VERY SLOW when the box is at or near capacity. Mark -- Mark Zelden Sr. Software and Systems Architect - z/OS Team Lead Zurich North America / Farmers Insurance Group - ZFUS G-ITO mailto:[EMAIL PROTECTED] z/OS Systems Programming expert at http://expertanswercenter.techtarget.com/ Mark's MVS Utilities: http://home.flash.net/~mzelden/mvsutil.html -- 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
Re: IPL an LPAR with a very low weight?
Hi Dave -- I suspect you wouldn't be considering this unless you were having problems with CPU resource contention in the two big LPARs (at 47 and 43). What kind of machine is this? how many engines? If this small LPAR is down, and the box is nearly 100% busy, IPLing this LPAR will be very slow, and this small LPAR may slow down the others in the sysplex and MIM. What is the real problem you are trying to solve? Al Al Sherkow, I/S Management Strategies, Ltd. Consulting Expertise on Capacity Planning, Performance Tuning, WLC, LPARs, IRD and LCS Software Seminars on IBM SW Pricing, LPARs, and IRD Voice: +1 414 332-3062 Web: www.sherkow.com -- 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