Re: How many IFLs on my box?
This was in my share presentation today. I will have to look into how standby processor information is provided or should be reported. This shows the lpars, their processor types, and then it shows the activity by processor type, so for the 37 active processors on this system, it should be obvious the utilization. Report: ESALPARS Logical Partition Summary Monitor initialized: 11/06/10 at 16:07:10 on 2097 serial 374E: 11/0 --- --Complex-- ---Logical Partition--- -Assi Proce Phys Dispatch Virt %Assigned ---LP Type Time CPUsSlice Name Nbr CPUs Total Ovhd Weight --- - -- - 16:09:0037 Dynamic Totals:0 50 3146 25.03000 L43 196 574.6 0.6 148 IFL C41 101 100.0 0.0 Ded ICF C42 111 96.1 0.1 850 ICF C43 141 99.7 0.0 Ded ICF C44 151 0.8 0.1 150 ICF P4117 422.1 3.2 717 CP P4492 43.4 0.2 70 CP T4145 197.5 0.5 193 CP T4472 9.8 0 20 CP L41 17 22 1557 19.6 777 IFL ?- 71% L42 182 44.7 0.8 75 IFL Totals by Processor type: -CPU--- -Shared Processor busy Type Count Ded shared total assigned Ovhd Mgmt - --- -- - CP 6 0 6 584.7573.3 3.6 7.8 IFL 27 0 27 2220 2176.3 21.0 22.9 ?-80% of IFLs ICF 3 0 3 297.8296.5 0.1 1.1 ZIIP 1 0 1 99.9 99.5 0.3 0.1 Marcy Cortes wrote: I failed to mention that I wanted it from z/VM or Linux programmatically :( I do have a nice spreadsheet from our cap folks who do the Resource Link thing. Marcy -Original Message- From: The IBM z/VM Operating System [mailto:IBMVM@LISTSERV.UARK.EDU] On Behalf Of Feller, Paul Sent: Tuesday, August 09, 2011 5:34 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: [IBMVM] How many IFLs on my box? If someone in your shop has access to the IBM website that I call Server Resource Link it can tell you what is installed in the box. The display below is from one of our z10-EC boxes. It has 3 IFLs installed but z/VM lpar on the box is only allowed to see 2 of them. Running CPs: 5 Running SAPs: 6 Running ICFs: 1 Running Linux: 3 Running zAAPs: 0 Running zIIPs: 2 Physical PUs: 34 CPs in LICCC: 5 SAPs in LICCC: 6 ICFs in LICCC: 1 Linux in LICCC: 3 zAAPs in LICCC: 0 zIIPs in LICCC: 2 Paul Feller AIT Mainframe Technical Support
Re: DASD utilization question
would you be looking for something from a really good performance monitor: Screen: ESAHST2 Velocity Software - VSIVM4 ESAMON 3.808 08/04 1 of 1 LINUX HOST Storage Analysis Report NODE R* LIMIT 500 -Utilization- -Storage-- MegaByte Pct Alloc Time Server Index Size Used Full Errors Units Description - - - -- - 07:07:00 RMTLINUX33 465 291 62.6 0 4096 /opt 329361 65.7 0 4096 / 31 372 334 89.9 0 1024 /Linux-s390 72323 100 0 1024 Cached memory 6 250 1 0.5 0 1024 Memory buffers 3 250 220 88.2 0 1024 Virtual memory 1 250 220 88.2 0 1024 Physical memory RH5Z3161 4 10904 2966 27.2 0 4096 / 3 409551 1.3 0 1024 Swap Space 2 497 493 99.1 0 1024 Real Memory 1 497 126 25.3 0 1024 Memory Buffers RH5Z2161 4 10904 2966 27.2 0 4096 / 3 409551 1.3 0 1024 Swap Space 2 497 493 99.1 0 1024 Real Memory 1 497 126 25.3 0 1024 Memory Buffers RH5Z161 4 10904 2966 27.2 0 4096 / 3 409551 1.3 0 1024 Swap Space 2 497 493 99.1 0 1024 Real Memory 1 497 126 25.3 0 1024 Memory Buffers Ticona, Luis wrote: Is there any tool that will allow me to display zVM DASD utilization. I am trying to get similar to the below information but from a zLINUX server running as a guest under zVM 5.4. q disk LABEL VDEV M STAT CYL TYPE BLKSZ FILES BLKS USED-(%) BLKS LEFT BLK TOTAL MNT191 191 A R/W 175 3390 4096 254 8934-28 22566 31500 MNT5E5 5E5 B R/W 9 3390 4096 131 1290-80 330 1620 MNT2CC 2CC C R/W 5 3390 4096 60407-45 493900 MNT51D 51D D R/W26 3390 4096 306 1575-34 3105 4680 MNT193 193 H R/W 167 3390 4096 1093 21035-70 9025 30060 MNT190 190 S R/O 100 3390 4096 691 14921-83 3079 18000 MNT19E 19E Y/S R/O 250 3390 4096 1021 28225-63 16775 45000 Ready; T=0.01/0.01 20:19:56 I go into one of our zLinux servers running under zVM 5.4 and did the same display command and the only thing I received is just Information about their label, vdev, mode. The stat column information was only OS. Didn’t get info for the files, used-(%), blks left and blk total columns. The volumes in this server are 3390-27. The last thing I will attempt to do is dump these dasd to tape using INNOVATION FDRABR and then get the information about the tracks being dumped from the small report after the JCL job output is completed. Thank you; *Luis Ticona* Management Information Systems Division 1 Police Plaza New York, NY 10038 ltic...@nypd.org or 646-610-5304 cid:image001.gif@01C911AB.7C344550
Re: Log usage of FTP
If they happen to have ZVPS (ESALPS), that information is captured using the TCPIP monitor records. Kris Buelens wrote: My customer want to see what IP addresses are still using FTP with his z/VM system. Is there a simpler way than coding a CHKIPADR exit exec, (or a NETSTAT every second like he does now)? I don't see a parameter one can code. Maybe some trace option? -- Kris Buelens, IBM Belgium, VM customer support
Re: VM question - QIDLE
I will take this opportunity to first agree with Dennis, (Thank you Dennis) and say that for VSE installations running under z/VM, we have a phenomenal offering called the ZVPS Starter Kit that won't break any even little banks. O'Brien, Dennis L wrote: Velocity Software has an ESAFORCE component which can force idle users. I imagine it could just list them, too. Dennis If you give a mouse a cookie, he's going to ask for a glass of milk. *From:* The IBM z/VM Operating System [mailto:IBMVM@LISTSERV.UARK.EDU] *On Behalf Of *Frank M. Ramaekers *Sent:* Monday, April 11, 2011 08:47 *To:* IBMVM@LISTSERV.UARK.EDU *Subject:* Re: [IBMVM] VM question - QIDLE Redirect… Frank M. Ramaekers Jr. *From:* owner-vs...@lehigh.edu [mailto:owner-vs...@lehigh.edu] *On Behalf Of *Wakser, David *Sent:* Monday, April 11, 2011 10:41 AM *To:* VSE Discussion List *Subject:* VM question - QIDLE Since our email addresses were changed, I have not been successful in re-subscribing to the VM discussion list, so I am posting this here. If someone would be so kind as to post this on the VM list, I would be greatly appreciative. We have a module in z/VM called QIDLE (dated 2002) which lists current CMS users and the number of minutes that they have been idle. It was originally written by Randy Foster when he did work for our company (or our client). However, we do not have the source, and Angela just informed me that Randy passed away quite a while ago. Does anyone have access to the source for such a program (that can be shared), or does anyone have another method of determining how many minutes a CMS user has been idle? Thanks, in advance. David Wakser (201) 840-4781 Confidentiality Note: This e-mail, including any attachment to it, may contain material that is confidential, proprietary, privileged and/or Protected Health Information, within the meaning of the regulations under the Health Insurance Portability Accountability Act as amended. If it is not clear that you are the intended recipient, you are hereby notified that you have received this transmittal in error, and any review, dissemination, distribution or copying of this e-mail, including any attachment to it, is strictly prohibited. If you have received this e-mail in error, please immediately return it to the sender and delete it from your system. Thank you. _ This message contains information which is privileged and confidential and is solely for the use of the intended recipient. If you are not the intended recipient, be aware that any review, disclosure, copying, distribution, or use of the contents of this message is strictly prohibited. If you have received this in error, please destroy it immediately and notify us at privacy...@ailife.com. This message w/attachments (message) is intended solely for the use of the intended recipient(s) and may contain information that is privileged, confidential or proprietary. If you are not an intended recipient, please notify the sender, and then please delete and destroy all copies and attachments, and be advised that any review or dissemination of, or the taking of any action in reliance on, the information contained in or attached to this message is prohibited. Unless specifically indicated, this message is not an offer to sell or a solicitation of any investment products or other financial product or service, an official confirmation of any transaction, or an official statement of Sender. Subject to applicable law, Sender may intercept, monitor, review and retain e-communications (EC) traveling through its networks/systems and may produce any such EC to regulators, law enforcement, in litigation and as required by law. The laws of the country of each sender/recipient may impact the handling of EC, and EC may be archived, supervised and produced in countries other than the country in which you are located. This message cannot be guaranteed to be secure or free of errors or viruses. References to Sender are references to any subsidiary of Bank of America Corporation. Securities and Insurance Products: * Are Not FDIC Insured * Are Not Bank Guaranteed * May Lose Value * Are Not a Bank Deposit * Are Not a Condition to Any Banking Service or Activity * Are Not Insured by Any Federal Government Agency. Attachments that are part of this EC may have additional important disclosures and disclaimers, which you should read. This message is subject to terms available at the following link: http://www.bankofamerica.com/emaildisclaimer. By messaging with Sender you consent to the foregoing.
Re: VM Performance Toolkit - Velocity Starter Kit
I might as well take this opportunity. Velocity Software has a new offering, for installations just getting started. The Starter Kit includes zVPS (the FULL Velocity Performance Suite) and zTUNE for a VERY low (not free) price. This provides the ability to completely monitor your z/VM, your Linux, your MS Servers, your Blade servers, and even your VSE servers. And we put someone on site for a day at no cost to make sure your starter kit gets started Hughes, Jim wrote: Is the VM Performance Toolkit a free product from IBM? Jim Hughes Consulting Systems Programmer Mainframe Technical Support Group Department of Information Technology State of New Hampshire 27 Hazen Drive Concord, NH 03301 603-271-5586Fax 603.271.1516 Statement of Confidentiality: The contents of this message are confidential. Any unauthorized disclosure, reproduction, use or dissemination (either whole or in part) is prohibited. If you are not the intended recipient of this message, please notify the sender immediately and delete the message from your system.
Re: cpuplugd Daemon
I would HIGHLY recommend talking to an installation who has actually implemented VMRM PRIOR to you doing it. I don't think even IBM recommends VMRM (ok maybe some sales types do). The problem is about your ability to keep your servers from not crashing if you care.. Dave Jones wrote: Hi, Rakesh. A good place to start learning about the z/.VM side of CMM is the z/VM V6R1 Performance document, available from the IBM z/VM online library. The VM side of CMM is implemented by the VMRMSVM virtual machine, which is already defined in the user directory. Chapter 4.6 in the Performance document above describes how this works in detail. DJ On 03/09/2011 06:50 AM, Rakesh Krishnakumar wrote: We are activating cpuplugd process for dynamic CPU and memory management for Linux guests running in z/VM. We have found a reference in Virtualization cook book for SLES11 SP1 how to make necessary configuration for CMM modules within Linux.But couldn't find any reference about the configuration to be done within z/VM for CMM. Is any configuration required or z/VM comes with CMM enabled by default. The z/VM version used is 6.1 and Linux is SLES11 SP1. Also does this process have any adverse performance implication? Regards Rakesh.K
Re: VM Total time in $ACCOUNT files
When i do the analysis for 100% capture ratio, i add the physical overhead, the logical overhead and the z/vm captured time. that works. so vm must be lpar aware... Gregg wrote: On Fri, Feb 18, 2011 at 7:31 PM, George Henke/NYLIC george_he...@newyorklife.com mailto:george_he...@newyorklife.com wrote: ...And so the important factor is no what % of phyiscal CP (PCP) bu what % of logical CPU is being realized in each LPAR, Guest Machine Some percent of an interval is going to be used by PR/SM for Physical CP and LPar Management. A small percentage but is it reflected in accounting records? If VM is watching the wall clock and the last time it looked at a machine it was running, PR/SM steps in for a couple of clicks and when VM gets back is the 'run time' the wall/TOD clock delta? Does VM know it wasn't running and where is that time accounted for or are only 'running' seconds counted? If the CEC is sufficiently busy so that the weights are in play and especially if the LPar is capped, the Total(PCP) number of CPU seconds available may be considerably less than the logical CPU seconds. It probably doesn't matter much at an individual virtual machine level but at an aggregate, I'm thinking the physical number may be more important than the logical. Gregg Reed No Plan, survives execution
Re: VM Total time in $ACCOUNT files
and for linux servers with multiple cpus, you divide cputime by the number of cpus? Ackerman, Derek wrote: Yes, and I always divide the CPU time by 1000. -Original Message- From: The IBM z/VM Operating System [mailto:IBMVM@LISTSERV.UARK.EDU] On Behalf Of Bruce Hayden Sent: Friday, February 18, 2011 10:14 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: VM Total time in $ACCOUNT files Hopefully you're converting the units. Virtual and Total time are in milliseconds and connect time is in seconds. Are you seeing this even after converting all of them to the same units? On Fri, Feb 18, 2011 at 1:04 PM, Ackerman, Derek derek.acker...@infocrossing.com wrote: I always thought that the total VM time, as well as the virtual, would be less than the connect time. Is it normal for the total time to be greater than the connect time? This happens a lot... Derek Ackerman Enterprise Capacity Planning Performance Management Infocrossing Inc Confidentiality Note: This e-mail, including any attachment to it, may contain material that is confidential, proprietary, privileged and/or Protected Health Information, within the meaning of the regulations under the Health Insurance Portability Accountability Act as amended. If it is not clear that you are the intended recipient, you are hereby notified that you have received this transmittal in error, and any review, dissemination, distribution or copying of this e-mail, including any attachment to it, is strictly prohibited. If you have received this e-mail in error, please immediately return it to the sender and delete it from your system. Thank you.
Re: VM Total time in $ACCOUNT files
how you do the arithmetic relates to the problem you are trying to solve Rich Greenberg wrote: On: Fri, Feb 18, 2011 at 11:05:07AM -0800,Barton Robinson Wrote: } and for linux servers with multiple cpus, you divide cputime by the } number of cpus? Wouldn't it be closer to reality to multiply the connect time by the number of cpus?
Re: SET SHARE ABSOLUTE/RELATIVE
Very few people understand the scheduler, and what it does - What it does, it does very well. You just have to understand the language. Rob did some recent experiments that validated how it works, and validated how little functions like VRM really help your workloads (except by accident). From the VelocitySoftware.com page, go to Linux Hints and Tips. There is a new link there to Rob's recent research. I would recommend reading it and understanding it before setting shares (or using VRM). Hughes, Jim wrote: Thanks for the reply Marty. Long time, no see. Our VSE systems are mainly interactive CICS or IDMS/DC systems during the day. Night time they become batch machines. The CICS and IDMS/DC systems are mainly accessed via VTAM. Our three production systems are each set to ABSOLUTE 20% with no defined target maximum. The sum of our ABSOLUTE SHARE users does total 100%. With that said, we’ve asked ourselves is ABSOLUTE 20% enough? The manual says once you have defined the minimum target ABSOLUTE SHARE to total 100%, the scheduler reserves 1% for the RELATIVE SHARE users. It goes on to say that once an ABSOLUTE SHARE user has reached its minimum target share it only gets more if system resources are available. What I am looking for is a way to keep the production systems behaving if a production vse system(absolute share), test vse system(relative share) or a cms user(relative share) begins to loop. The more I read about CP SET SHARE the more I suspect it isn’t designed to be a panacea for smooth performance in time of trouble. Maybe I should be investigating the VM Performance Monitor to assist with dynamic performance adjustment in a time of trouble. Comments? Jim Hughes Consulting Systems Programmer Mainframe Technical Support Group Department of Information Technology State of New Hampshire 27 Hazen Drive Concord, NH 03301 603-271-5586Fax 603.271.1516 Statement of Confidentiality: The contents of this message are confidential. Any unauthorized disclosure, reproduction, use or dissemination (either whole or in part) is prohibited. If you are not the intended recipient of this message, please notify the sender immediately and delete the message from your system. *From:* The IBM z/VM Operating System [mailto:IBMVM@LISTSERV.UARK.EDU] *On Behalf Of *Martin Zimelis *Sent:* Monday, February 07, 2011 3:01 PM *To:* IBMVM@LISTSERV.UARK.EDU *Subject:* Re: SET SHARE ABSOLUTE/RELATIVE Catherine, I don't think your understanding of SHARE is backwards, but your expectation of what the performance manager will do might be. I suspect it's trying to keep heavy CPU users from hogging the processors. To get back to the original question, Jim, I think you need to describe what the z/VSE guests are doing. If they're supporting interactive users (e.g., CICS), you'd want one answer from the assembled masses. If they're true batch workloads, the answer should be quite different. Since your system's perceived responsiveness likely depends on how quickly TCPIP (and VTAM) gets serviced, a high share is called for. In your situation, is the same true for RSCS? Regardless, my experience with the conventional wisdom of whether to use relative or absolute shares is dated, so I'll leave detailed recommendations to those with more recent experience. Marty Zimelis On Mon, Feb 7, 2011 at 2:13 PM, McBride, Catherine cmcbr...@kable.com mailto:cmcbr...@kable.com wrote: A while ago a very experienced VM person from IBM suggested that we not use ABSOLUTE unless you absolutely must cap off a guest to keep it from running away with your real processors. We used that setting on our test system only. Our VSE TOR and VM guest TCPIP both had high relative shares (1 versus 3000 for regular production guests). Then we started using a performance manager feature of VM Toolkit, it managed share values for us. It set everything the same after VM IPL, but by the end of a normal production day our busiest guests had dropped to the lowest relative share, the ones seldom used had the highest. Meaning my understanding of how relative share worked was backwards or the gizmo in VM Toolkit was. Hopefully Alan or Kris will expound. -Original Message- From: The IBM z/VM Operating System [mailto:IBMVM@LISTSERV.UARK.EDU mailto:IBMVM@LISTSERV.UARK.EDU] On Behalf Of Hughes, Jim Sent: Monday, February 07, 2011 12:57 PM To: IBMVM@LISTSERV.UARK.EDU mailto:IBMVM@LISTSERV.UARK.EDU Subject: SET SHARE ABSOLUTE/RELATIVE I've read the CP COMMAND manual and the PERFORMANCE manual regarding the SET SHARE command and how it works. Would someone care to comment on how you have used them for your z/VSE production and guest machines? What would suggest for TCPIP/RSCS/VTAM SET SHARE values? Thanks
Performance Education scheduled
For those of you interested in performance, or performance management of your Linux and z/VM environment, our Performance Education schedule is now posted for 1st half, at http://velocitysoftware.com/seminar/index.html;. Offerings are the 4 day workshop and 1 day seminars.
Re: Strange response time problems (also posted on VSE-L)
so i know sales people that could help you solve your problem Wouldn't it be nice to tell your boss that you know what happened, it is fixed and won't happen again? Wakser, David wrote: I’ll have to wait until it happens again! J *From:* The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] *On Behalf Of *Neale Ferguson *Sent:* Tuesday, December 21, 2010 12:21 PM *To:* IBMVM@LISTSERV.UARK.EDU *Subject:* Re: Strange response time problems (also posted on VSE-L) Actually, what does #CP IND Q show when the problem is occuring? On 12/21/10 12:17 PM, Quay, Jonathan (IHG) jonathan.q...@ihg.com wrote: Guest operating systems almost always live in Q3. Try bumping up the Q3 STORBUF. *From:* The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] *On Behalf Of *Wakser, David *Sent:* Tuesday, December 21, 2010 10:21 AM *To:* IBMVM@LISTSERV.UARK.EDU *Subject:* Re: Strange response time problems (also posted on VSE-L) Already did that: q srm IABIAS : INTENSITY=90%; DURATION=3 LDUBUF : Q1=300% Q2=200% Q3=150% STORBUF: Q1=300% Q2=200% Q3=150% DSPBUF : Q1=32767 Q2=32767 Q3=32767 DISPATCHING MINOR TIMESLICE = 5 MS MAXWSS : LIMIT=% .. : PAGES=99 XSTORE : 0% *From:* The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] *On Behalf Of *Helmuth Teubl *Sent:* Tuesday, December 21, 2010 10:06 AM *To:* IBMVM@LISTSERV.UARK.EDU *Subject:* Strange response time problems (also posted on VSE-L) Hi, maybe SRM-Settings not OK? You should overcommit LDUBUFs and STORBUFs The defaults are: CP Q SRM IABIAS : INTENSITY=90%; DURATION=2 LDUBUF : Q1=100% Q2=75% Q3=60% STORBUF: Q1=125% Q2=105% Q3=95% DSPBUF : Q1=32767 Q2=32767 Q3=32767 DISPATCHING MINOR TIMESLICE = 5 MS MAXWSS : LIMIT=% .. : PAGES=99 XSTORE : 0% LIMITHARD METHOD: DEADLINE Have a look, maybe try following settings: q srm IABIAS : INTENSITY=90%; DURATION=2 LDUBUF : Q1=300% Q2=200% Q3=100% STORBUF: Q1=300% Q2=275% Q3=250% DSPBUF : Q1=32767 Q2=32767 Q3=32767 DISPATCHING MINOR TIMESLICE = 5 MS MAXWSS : LIMIT=% .. : PAGES=99 XSTORE : 0% LIMITHARD METHOD: DEADLINE kind regards Helmuth David.Wakser---21.12.2010 15:54:15---All: We are running 2 2.3 VSE systems under z/VM 5.4 on a Z800 CPU. Von: david.wak...@infocrossing.com An: IBMVM@LISTSERV.UARK.EDU Datum: 21.12.2010 15:54 Betreff: Strange response time problems (also posted on VSE-L) Gesendet von: IBMVM@LISTSERV.UARK.EDU All: We are running 2 2.3 VSE systems under z/VM 5.4 on a Z800 CPU. We are experiencing periods of time when VSEs do not respond at all (e.g. cannot get in via FAQS from CMS, etc.), even though nothing is running in the VSE system and the z/VM system is not very busy. At other times, we have excellent response times, though conditions on either the guests or the z/VM host didn’t seem to change. We do not have any z/VM monitors (except Explore, which is not set up properly), and we have even tried QUICKDSP, without success. Is anyone aware of any PTFs that address this strange behavior? We believe it started when the system was upgraded to z/VM 5.4. David Wakser Confidentiality Note: This e-mail, including any attachment to it, may contain material that is confidential, proprietary, privileged and/or Protected Health Information, within the meaning of the regulations under the Health Insurance Portability Accountability Act as amended. If it is not clear that you are the intended recipient, you are hereby notified that you have received this transmittal in error, and any review, dissemination, distribution or copying of this e-mail, including any attachment to it, is strictly prohibited. If you have received this e-mail in error, please immediately return it to the sender and delete it from your system. Thank you. Raiffeisen Informatik GmbH, Firmenbuchnr. 88239p, Handelsgericht Wien, DVR 0486809, UID ATU 16351908 Der Austausch von Nachrichten mit oben angefuehrtem Absender via E-Mail dient ausschliesslich Informationszwecken. Rechtsgeschaeftliche Erklaerungen duerfen ueber dieses Medium nicht ausgetauscht werden. Correspondence with above mentioned sender via e-mail is only for information purposes. This medium may not be used for exchange of legally-binding communications. Confidentiality Note: This e-mail, including any attachment to it, may contain material that is confidential, proprietary, privileged and/or Protected Health Information, within the meaning of the regulations
Re: Hanging when dialing z/OS guests
And of course if you look at zpro, it does most of that as well as a free option of a VERY inexpensive add on. Tracy Dean wrote: Since Mike mentioned CA's VM:Spool, I'll add my obligatory mention of IBM Operations Manager for z/VM. It lets you view a list of files in the spool (with owner, size, date details), filter and sort the list, look at the contents of spool files, display a list of the top ten largest spool files, and display a list of the top ten users with the most spool files. Tracy Dean IBM
Re: The old VM/ESA CMS GUI - Does it still live?
I have multiple installations running Linux under z/VM where the z/VM sysprogs are very overcommitted. Allowing them to offload work that CAN be done by someone else makes them much more productive - as long as the work assigned is controlled by an authorization mechanism. Most of the work, even DNS admin functions are janitor work, or maybe even clerical (or management), why utilize a skilled person for those functions? But yes, every installation needs access to someone with skills, and I'd much rather have 10 new installations, sharing one sysprog than zero with great skills... William D Carroll wrote: I think we can safely assume that since this discussion was about VM CMS that we're talking operating systems and interfaces to those OS's not applications as is your wife's case GUI's have their place, I personally do not believe that place is in an OS to manage it or it's subsystems where understanding what happens is very important. I know DNS admins who can't manage their servers without GUI's. That to me is ridiculous and are admins who need to understand more about what they manage. Are they more productive maybe, are they dangerous, I leave that to you Think of this, what happens to those admins if their gui breaks and all they have is command line. William 'Doug' Carroll -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Schuh, Richard Sent: Wednesday, November 24, 2010 12:54 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: The old VM/ESA CMS GUI - Does it still live? That is an assumption or opinion, not necessarily a fact. For example, my wife may not know the operating system line mode commands, but she certainly knows the accounting systems that she uses via a GUI interface. In fact, the GUI represents an increase in productivity. I would suggest that if you do not know what you are doing, then you should not use either interface; if you do know, you should use whatever interface you choose. Rule Number 1 (Know what you are doing) applies. Regards, Richard Schuh You still type in your 3270 emulator, you still have to know what you're doing. GUI's hide that from you Doug This communication is for informational purposes only. It is not intended as an offer or solicitation for the purchase or sale of any financial instrument or as an official confirmation of any transaction. All market prices, data and other information are not warranted as to completeness or accuracy and are subject to change without notice. Any comments or statements made herein do not necessarily reflect those of JPMorgan Chase Co., its subsidiaries and affiliates. This transmission may contain information that is privileged, confidential, legally privileged, and/or exempt from disclosure under applicable law. If you are not the intended recipient, you are hereby notified that any disclosure, copying, distribution, or use of the information contained herein (including any reliance thereon) is STRICTLY PROHIBITED. Although this transmission and any attachments 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 JPMorgan Chase Co., its subsidiaries and affiliates, as applicable, for any loss or damage arising in any way from its use. If you received this transmission in error, please immediately contact the sender and destroy the material in its entirety, whether in electronic or hard copy format. Thank you. Please refer to http://www.jpmorgan.com/pages/disclosures for disclosures relating to European legal entities.
Re: DASD Inventory
If you look at zpro (live system running many linux servers), at vm2.velocitysoftware.com/zpro, sign in with the userid and password provided, go to System status then dasd, gets you our live dasd report. This user is of course restricted, otherwise y'all would be seeing how bad you could hurt us. but click on an address of a device with a few minidisks, gives you an idea of what you can do. Sorry, we don't do z/OS, (but we DO know how...) Ticona, Luis wrote: Good day! Can somebody tell me if there is a good way or tool outhere to help us manage zVM and zOS dasd usage and allocation. Also creating reports and charts. Our shop is constantly growing with Linux under zVM 5.4. Your suggestions will be greatly appreciated. Thank you! *Luis Ticona* Management Information Systems Division 1 Police Plaza New York, NY 10038 ltic...@nypd.org or 646-610-5304 cid:image001.gif@01C911AB.7C344550
Re: The old VM/ESA CMS GUI - Does it still live?
Ok, Sorry George, I (who can type) find that offensive. The biggest challenge the z/vm platform has to grow is lack of people that understand 3270, or are even interested. Or do y'all want the platform to die as us people that can type get old, retire? vmware is winning the battle world wide with their gui and now are the base line for virtualization. So we either conform to how the world has changed, or be the dinasour they think we are. If you look at http://vm2.velocitysoftware.com/zpro;, this is our demo system of what we're going to ship as ZPRO. ZPRO started as a simple cloning tool, but now has turned into a project to address z/VM's real needs, hoping to attract the minds and hearts of future generations. George Henke/NYLIC wrote: Please tell ur laughing friends that GUI, *point and click*, is for people who can't type. In the mainframe world and particularly z/VM you not only need to know how to *type* but also *think* the old shibboleth of IBM, which made it great, and which George Bernard Shaw claimed occurs among us only once or twice a year, though he claimed he could think once a week. GUI is *smoke and mirrors*, dream stuff. OTOH, Mainframe, *Green Screen*, or whatever is a real operating system which does the thinking, really processing, since computers can't think, behind the GUI curtain. *Jeff Gribbin jeff.grib...@gmail.com* Sent by: The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU 11/23/2010 07:54 AM Please respond to The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU To IBMVM@LISTSERV.UARK.EDU cc Subject The old VM/ESA CMS GUI - Does it still live? Greetings folks, Some recent discussion brought to mind the old CMS GUI Facility that ship ped with VM/ESA 2.1 - and now I'm in a position where I might for the first t ime actually be able to configure a z/VM system to allow me to play with the beast, I got to wondering if it even still exists. The best I've been able to find in the way of IBM documentation is: http://www.vm.ibm.com/gui/ I found an interesting reference to a 2004 z/VM GUI project - http://web2.clarkson.edu/projects/cosi/zTeam/zvmgui/ but alas that seems to have since sunk without trace. So ... a few questions ... Anybody using it? Anybody prepared to admit they're using it? Anybody know if it's still maintained in any current form? This is purely a personal learning exercise triggered by the aforemention ed conversation and my recent exposure to young sysprogs ( 25 years old) wh o tend to love what CP and CMS can do but fall about laughing whenever the user interface is discussed. All comments gratefully received. Regards Jeff
Re: VM Monitor data processing
Ouch. Pretty costly to collect raw data, ship it to z/os and process it there. There are lots of installations that take the zVPS (Velocity Performance Suite) data and ship it to z/os from vm, and MXG supports it direct. Probably 1 percent in size. Other installations take our MICS file (probably less than 1 percent in size) and ship that to z/OS for processing there. Other vendors are in process to take the data as well for the same reasons (plus the get linux and other network data). Anything to stop collecting raw data Ackerman, Derek wrote: Is anyone transmitting to an MVS host the VM Monitor data to be processed by MXG/SAS programs? Derek Ackerman Enterprise Capacity Planning Performance Management Infocrossing Inc (206) 432-9737 || C: (206) 225-3585 || derek.acker...@infocrossing.com Please consider the environment before printing this e-mail. Confidentiality Note: This e-mail, including any attachment to it, may contain material that is confidential, proprietary, privileged and/or Protected Health Information, within the meaning of the regulations under the Health Insurance Portability Accountability Act as amended. If it is not clear that you are the intended recipient, you are hereby notified that you have received this transmittal in error, and any review, dissemination, distribution or copying of this e-mail, including any attachment to it, is strictly prohibited. If you have received this e-mail in error, please immediately return it to the sender and delete it from your system. Thank you.
Re: z/VM ISFC links
Isn't it absolutely unbelievably amazing that in the current environment with everything internet enabled, that z/vm is still stuck with 30 year old technology (CTC) to perform simple network functions? With no change anywhere in the future? *$# unbelievable Mark Pace wrote: I see that now. 1st criteria for this test is to share SFS across LPARs. 2nd was to start learning about what will be involved with SSI. So I guess I'm sticking to ISFC. Glad I have extra ESCON and FICON CHPIDs. Guess I'll start with ESCON as I also have extra cables, no extra FICON cables. On Thu, Sep 30, 2010 at 12:49 PM, Rob van der Heij rvdh...@gmail.com mailto:rvdh...@gmail.com wrote: On Thu, Sep 30, 2010 at 6:21 PM, Mark Pace pacemainl...@gmail.com mailto:pacemainl...@gmail.com wrote: I think I'll also look into IPGATE. But that does not do ISFC ... -- Mark D Pace Senior Systems Engineer Mainline Information Systems
Re: z/VM ISFC links
My internal network doesn't need encryption, and i should be able use hypersockets between LPARs. it would be better if you insist on having ctcs that the ctc are connected to tcpip as another interface. you simplify our options, tcpip will always be there, less complexity for those of us that like simple, and probably less money for those that care. Alan Altmark wrote: On Thursday, 09/30/2010 at 02:00 EDT, Barton Robinson bar...@vm1.velocity-software.com wrote: Isn't it absolutely unbelievably amazing that in the current environment with everything internet enabled, that z/vm is still stuck with 30 year old technology (CTC) to perform simple network functions? With no change anywhere in the future? *$# unbelievable No change anywhere in the future? Who said that? And, btw, IBM has actually given some thought to the problem. If you want to go over ethernets, then you're going to be dealing with IP connectivity *and encryption*, and all that entails. I don't think the VM IP stack is up to the challenge of pushing that data, so alternatives are needed. Personally, I think I'd rather have FICON for now. Ethernet was invented in the early 70s, so the 30-year-old tech argument applies to both. Alan Altmark z/VM and Linux on System z Consultant IBM System Lab Services and Training ibm.com/systems/services/labservices office: 607.429.3323 alan_altm...@us.ibm.com IBM Endicott
Re: Is there a JAVA implementation for CMS?
Right, so even better chance that java on cms would be more than acceptable. Too bad there's no business case. John P. Hartmann wrote: Barton, you overlook the lack of IEEE floating point in the hardware in those days. On 28 September 2010 04:08, Barton Robinson bar...@vm1.velocity-software.com wrote:
Re: Is there a JAVA implementation for CMS?
I think the cms java performed very poorly could be revisited. back then, the mainframe was at most 400Mhz? or less? of course it performed poorly. so now the processors have caught up and surpassed and are about an order of magnitude faster. This had nothing to do with software, this was a hardware problem that was very poorly understood. As i've said before, way back then we thought the mainframe was big and fast, and we were half right. Alan Ackerman wrote: Unfortunately, CMS Java performed very poorly so I would not recommend try to find a copy. I think people waited for IBM to fix the Java performance, and IBM waited for customers to use it, and eventually it went away. I do not remember any screaming on this list when the removal was first announced. I was hoping we would get Java Servlets, but I never heard of any web server vendor offering to support them. And no, I didn't open a PMR on the poor performance. I wish that I had. Did anyone? In any case, this is only one of the many ways IBM (and vendors) have not continued to provide modern application function for CMS. (XML, web services, etc.) It breaks my heart, but I have to recommend people use Linux instead of CMS. Without vendor interest I cannot recommend CMS for any new function. Old applications will continue to run, as IBM is still supporting most CMS functions. At our shop, when they seek to add new functions (I wanted to add web services) they have to convert off of CMS. Unfortunately, that means converting off of System z here. I'm trying to reverse that by working on z/Linux Support, but it's too late for all but a few of our CMS applications. Alan Ackerman, Bank of America.com On Mon, 27 Sep 2010 11:02:32 -0400, Michael Donovan dono...@us.ibm.com wrote: There was a Java port for CMS, in days long gone. It was withdrawn from the product and the web pages were removed after z/VM 4.4.0. If you had installed the Developer Release 1.1.6 on a system prior to that and migrated that Byte File System forward, you could continue to use it on newer z/VM releases, even though it is no longer supported. From: Dave Jones d...@vsoft-software.com To: IBMVM@LISTSERV.UARK.EDU Date: 09/27/2010 10:56 AM Subject:Re: Is there a JAVA implementation for CMS? Sent by:The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU Yes, there is, but it's very old, and certainly out of date. You might try logging onto the z/VM home page and searching for IBM® JavaTM Port for VM/ESA, Developer Release 1.1.6. Good luck. On 09/27/2010 09:50 AM, Michel Beaulieu wrote: Hello, I want to share JAVA code between platforms. Is there a JAVA implementation for CMS? Thanks, Michel Beaulieu Montreal, Canada |*| -- Dave Jones V/Soft Software www.vsoft-software.com Houston, TX 281.578.7544
Re: CP unresponsive on certain guests
Gee Mike, you come to my class and don't learn nothen. the only thing that storbuf does is hurt, turning it off is the only recommendation that anyone ever gives. i look at this thread and So, anyone reading this, IF YOU HAVE A PERFORMANCE PROBLEM, if you send us z/vm monitor data, we will analyze it for you FOR FREE, everybody is guessing, and it really is a whole lot easier if you just collect a few minutes of data and send it to us. We have this really cool tool (ztune) we run the data through to give a full configuration check, health check, and performance analysis. Once we have the data, it's like 2 minutes to get this report. Can we make it any easier? Mike At HammockTree wrote: Yeah, that is probably where he needs to end up Dave, but I'm a little hesitant to recommend the 300% for Q3 without feeling more comfortable about his paging subsystem... Moving a couple of large guests from the E-list to in-Q could cause a increase in paging that he may or may not be configured to handle. Mike - Original Message - From: Dave Jones d...@vsoft-software.com To: IBMVM@LISTSERV.UARK.EDU Sent: Thursday, September 16, 2010 11:08 AM Subject: Re: CP unresponsive on certain guests Actually, Mike, he may be better off (a bit, at least) by setting STORBUFF 300 300 300. On 09/16/2010 09:58 AM, Mike At HammockTree wrote: Since the STORBUF setting is exactly the values I suggested, I suspect you applied the SET SRM STORBUFF 300% 250% 200% prior to doing the Q SRM With the current setting for STORBUFF, are you still experiencing the problem? Also, on a related note, what does your zVM paging system look like? The output of CP Q ALLOC PAGE will provide the information Mike - Original Message - From: Daniel Tate daniel.t...@gmail.com To: IBMVM@LISTSERV.UARK.EDU Sent: Thursday, September 16, 2010 10:52 AM Subject: Re: CP unresponsive on certain guests Output of Q SRM q srm IABIAS : INTENSITY=90%; DURATION=2 LDUBUF : Q1=100% Q2=75% Q3=60% STORBUF: Q1=300% Q2=250% Q3=200% DSPBUF : Q1=32767 Q2=32767 Q3=32767 DISPATCHING MINOR TIMESLICE = 5 MS MAXWSS : LIMIT=% .. : PAGES=99 XSTORE : 0% Ready; T=0.01/0.01 09:49:05 On Wed, Sep 15, 2010 at 5:47 PM, Dave Jones d...@vsoft-software.com wrote: Hi, Daniel. The answer to your first question is to use the CP FORCE command (HELP CP FORCE will tell you all about it.) The VM user id issuing the FORCE command needs to have privilege class A as well. Usually this is done from either MAINT or OPERATOR. The answer to your second question is a bit more difficult, I'm afraid. As Marcy has already suggested, what does a Q SRM command show? My first guess would be that your SLES11 guest is falling into Q3 and never given an opportunity to run. To find out *why* the guest is not able to run, you need the services of a good z/VM performance monitor.IBM offers the Performance Monitor (it comes bundles with z/VM, but it's an extra cost offering) and Velocity Software (http://www.velocity-software.com/) has a very good suite of products as well. IMHO it' practically impossible to run a modern production grade z/VM-zLinux system without a good performance monitor to help solve issues like the one your having now. On 09/15/2010 05:14 PM, Daniel Tate wrote: We're starting to run apps on the servers now. From time to time a guest will become unresponsive - to be more precise, ,the CP will not respond to commands, and neither will the guest OS (SLES11). not even #CP LOGOFF is acknowledged. from another login, CP INDIIC LOAD shows no appreciable load. Two questions from this: 1) how would I force a logoff of a user from another user? Is this possible? 2) if we are not paging and the IFLs are not loaded (2-3% utilization as a matter of fact) what could the bottleneck be? -- Dave Jones V/Soft Software www.vsoft-software.com Houston, TX 281.578.7544 -- Dave Jones V/Soft Software www.vsoft-software.com Houston, TX 281.578.7544
Re: Updates to requirements WAVV201007, WAVV201012
hehehehehehehe - good performance tools take care of this problem David Boyes wrote: Forwarded without comment. User Group Number - WAVV201012 Document Status - Recognized Title - MONDCSS is too small as shipped IBM agrees with the request and a solution appears to be a desirable objective. A solution however may not presently appear feasible or implementable. No IBM commitment is made or implied as to the eventual delivery of an acceptable solution.
Re: MONDCSS segment update
Totally safe. Only users of the star monitor function will notice - so the monitor does need to stop and start. Dean, David (I/S) wrote: Is it safe to change MONDCSS sizes on running system, i.e., I won’t run over memory being used somewhere else? David M. Dean Information Systems BlueCross BlueShield Tennnessee - Please see the following link for the BlueCross BlueShield of Tennessee E-mail disclaimer: http://www.bcbst.com/email_disclaimer.shtm
Re: Automated Logoff of CMS user
Have you looked at the TUNEFRC function, part of zVPS (Velocity Performance Suite)? Martin, Terry R. (CMS/CTR) (CTR) wrote: Hi, This may have been asked before but I was wondering the best way to Automatically log off a CMS user after a designated time frame. This is to address an Audit finding. Thank! /Thank You,/ / / /Terry Martin/ /Lockheed Martin - Citic/ /z/OS and z/VM Performance Tuning and Operating Systems Support/ /Office - 443 348-2102/ /Cell - 443 632-4191/ / / /cid:image001.jpg@01C97FB5.5EAFD6C0///
Re: Linux / Websphere memory creep
Gee, if you could send me some ESA reports, I'd be happy to review it. Dean, David (I/S) wrote: zLinux 5.4, SUSE 10.2. Running one instance of WebSphere 7 with 6 profiles / JVM instances. I started with 3 GIG, went to 5GIG, now at 6 GIG, and free memory still drops little by little, SWAP eventually begins to grow and finally performance goes to heck. Just like Windows (ouch) you reboot and everything is cool for a few more days…. I have read and heard the lectures on zLinux taking all you give it, but I need it to STOP. Yesterday procs ---memory-- ---swap-- -io -system-- -cpu-- r b swpd free buff cache si sobibo in cs us sy id wa st 0 0 0 140192 273624 129759600 0 8 2850 241 1 0 99 0 0 0 0 0 139448 273668 129755200 0 9 2254 249 1 0 99 0 0 0 0 0 139076 273692 129752800 0 9 2235 245 1 0 99 0 0 Today procs ---memory-- ---swap-- -io -system-- -cpu-- r b swpd free buff cache si sobibo in cs us sy id wa st 0 0 0 46300 292004 116002000 327 1783 269 3 0 94 0 3 0 0 0 46304 292064 115996000 214 2398 257 2 0 97 0 0 0 0 0 45668 292164 116140800 046 1832 287 4 0 94 0 1 0 0 0 45420 292228 116134400 016 2635 269 1 0 97 0 1 Here is the page info from Perfkit. DASD page creeps up just as memory. _Userid_ http://10.30.7.11:81/01BD5EB8/6E72/HE.04.001_Owned_ http://10.30.7.11:81/01BD5EB8/6E72/HE.04.010 _Reads_ http://10.30.7.11:81/01BD5EB8/6E72/HE.04.018 _Write_ http://10.30.7.11:81/01BD5EB8/6E72/HE.04.024 _Steals_ http://10.30.7.11:81/01BD5EB8/6E72/HE.04.030 _2GB_ http://10.30.7.11:81/01BD5EB8/6E72/HE.04.038 _XMS_ http://10.30.7.11:81/01BD5EB8/6E72/HE.04.044 _MSX_ http://10.30.7.11:81/01BD5EB8/6E72/HE.04.049 _XDS_ http://10.30.7.11:81/01BD5EB8/6E72/HE.04.054_WSS_ http://10.30.7.11:81/01BD5EB8/6E72/HE.04.062 _Resrvd_ http://10.30.7.11:81/01BD5EB8/6E72/HE.04.066 _R2GB_ http://10.30.7.11:81/01BD5EB8/6E72/HE.04.074 _R2GB_ http://10.30.7.11:81/01BD5EB8/6E72/HE.04.081 _L2GB_ http://10.30.7.11:81/01BD5EB8/6E72/HE.04.087 _L2GB_ http://10.30.7.11:81/01BD5EB8/6E72/HE.04.094 _XSTOR_ http://10.30.7.11:81/01BD5EB8/6E72/HE.04.101 _DASD_ http://10.30.7.11:81/01BD5EB8/6E72/HE.04.109 _Size_ http://10.30.7.11:81/01BD5EB8/6E72/HE.04.116 _Users_ http://10.30.7.11:81/01BD5EB8/6E72/HE.04.122 _LNX086_ http://10.30.7.11:81/01BD5EB8/6E72/USE.LNX086 .0 70.2 54.3 .0 .0 14.9 67.6 50.1 615164 0 71788 542585 4 40 69580 1134k 6144M Rob? Barton? Come on. David M. Dean Information Systems BlueCross BlueShield Tennnessee - Please see the following link for the BlueCross BlueShield of Tennessee E-mail disclaimer: http://www.bcbst.com/email_disclaimer.shtm
Re: V-disk
I've looked at a lot of installations using vdisk - I've only once seen an installation where vdisk seemed to be an issue - and that was when the installation had followed the 2 vdisk for swap guideline, but reversed the priority on the two disks and the larger one was used instead of the small one. So unless you've come up with a new imaginative way to use vdisk, use INFINITE and use the ESAMAP report ESASTR1 to understand the cost. There are many more important things to work on Schuh, Richard wrote: Is there a formula that can be used to determine what the default SYSLIM for V-disks would be in the absence of any SYSTEM CONFIG definitions? If I enter QUERY FRAMES, I get this: q frames All Frames: Configured=38273023 Real=38273023 Usable=38273023 Offline=0 Pageable=37871142 NotInitialized=0 GlobalClearedAvail=160 LocalClearedAvail=160 LocalUnclearedAvail=155 Frames 2G: GlobalUnclearedAvail=56339 Pageable=513876 LogicalFreeStorage=5602 RealFreeStorage=5 LockedRS=497 LockedCmd=0 MinidiskCache=8 Nucleus/Prefix=2521 Trace=400 Other=1386 Frames 2G: GlobalUnclearedAvail=4046728 Pageable=37357266 LogicalFreeStorage=55733 RealFreeStorage=19 LockedRS=15358 LockedCmd=0 MinidiskCache=97600 Other=320360 Assuming that the dasd space is sufficient to not be a limiting factor, does this give enough information to calculate a default SYSLIM? Regards, Richard Schuh
Re: ACM award - they deserve it....
If you go to conferences such as CMG (Computer Management Group), that has been a mainframe organization (meaning MVS or z/OS) since it started, our VM has never been represented, but VMWare now has many sessions. It's depressing to see 80 people in entry level performance session for VMWare and no z/VM sessions on the agenda of a mainframe conference. Early this year I was hearing ads for VMWare on the local radio station. I can only assume that VM is being outmarketed worldwide (or at least that VMWare is being marketed worldwide and VM is not marketed publicly at all). It doesn't matter if our mousetrap is better if nobody is out there trying to get mindshare (marketing). Preaching/grumbling to the choir doesn't change anything. So when was the last time that any of you tried to get a case study published showing how great your accomplishments are using z/VM? There are very few published stories (sorry games on z don't impress bean counters or executives, it's rather demeaning), we need REAL business case studies showing the value of z/VM to real companies. If we get enough and executives do a google search on VM, maybe they will find something useful? Bill Munson wrote: Jim, You are right, that makes me mad also. IBM really blew it when they did not trade mark VM munson Jim Elliott jelli...@gdlvm7.vnet.ibm.com Sent by: The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU 03/30/2010 09:34 PM Please respond to The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU To IBMVM@LISTSERV.UARK.EDU cc Subject Re: ACM award Today the Association for Computing Machinery (of which I have been a member since 1970) made the following award: VMware Workstation 1.0, the Software System Award, for bringing virtualization technology to modern computing environments, spurring a shift to virtual-machine architectures, and allowing users to efficiently run multiple operating systems on their desktops. Aside from the run multiple OSes on the desktop part, shouldn't we be insulted? Chip: Yes, we should be insulted. I remember being very upset the first time I heard a VMware employee talk about how they had invented the idea of server virtualization! Even on x86, VM386 was out years before VMware (even if it failed in the market). I am still upset every time I hear someone talk about VM when they mean VMware. My reaction is, I work on the real VM! Jim (aka Sir Jim the Evangelist) *** IMPORTANT NOTE*-- The opinions expressed in this message and/or any attachments are those of the author and not necessarily those of Brown Brothers Harriman Co., its subsidiaries and affiliates (BBH). There is no guarantee that this message is either private or confidential, and it may have been altered by unauthorized sources without your or our knowledge. Nothing in the message is capable or intended to create any legally binding obligations on either party and it is not intended to provide legal advice. BBH accepts no responsibility for loss or damage from its use, including damage from virus.
acm/vmware
The listserv sent me a message my post didn't go out, so try again. If you go to conferences such as CMG (Computer Management Group), that has been a mainframe organization (meaning MVS or z/OS) since it started, our VM has never been represented, but VMWare now has many sessions. It's depressing to see 80 people in entry level performance session for VMWare and no z/VM sessions on the agenda of a mainframe conference. Early this year I was hearing ads for VMWare on the local radio station. I can only assume that VM is being outmarketed worldwide (or at least that VMWare is being marketed worldwide and VM is not marketed publicly at all). It doesn't matter if our mousetrap is better if nobody is out there trying to get mindshare (marketing). Preaching/grumbling to the choir doesn't change anything. So when was the last time that any of you tried to get a case study published showing how great your accomplishments are using z/VM? There are very few published stories (sorry games on z don't impress bean counters or executives, it's rather demeaning), we need REAL business case studies showing the value of z/VM to real companies. If we get enough and executives do a google search on VM, maybe they will find something useful? There are many places to post and publish. Even twitter or blogs would be helpful in getting mindshare.
Re: [?? Probable Spam] Re: Perfkit SAMPLE CONFIG size too small
very large mondcss segments do not impact performance, only small ones do. B On Mar 30, 2010, at 11:21, RPN01 nix.rob...@mayo.edu wrote: Our MONDCSS grew, perhaps too large, while fighting this type message a long time ago. Once the problem was resolved, we didn't attempt to back off the changes we'd made, and the large size doesn't seem to hurt anything at the moment. I know that ultimately, making the segment larger was not the answer to the problem at the time, either. Also, Mr. Nunsford says hello. -- Robert P. Nix Mayo Foundation.~. RO-OE-5-55 200 First Street SW/V\ 507-284-0844 Rochester, MN 55905 /( )\ -^^-^^ In theory, theory and practice are the same, but in practice, theory and practice are different. On 3/30/10 12:57 PM, Eginhard Jaeger e.jae...@ch.inter.net wrote: There is no single 'right' MONDCSS size for all systems: it's about performance, so 'it depends'. The MONDCSS has to be large enough to allow the CP monitor to place all the monitor records you told it to collect in that storage area. Since most users just go and enable whole domains, it's the domains generating the largest number of monitor records that one wants to watch. For sample records that is, on most systems, the I/O domain, where you could end up with tens of thousands of devices already years ago when I still worked with VM. Be aware that the monitor will create a device activity record 3 of 268 bytes and a cache activity record 4 of 264 bytes for each DASD, and they must all fit simultaneously into the MONDCSS, together with all the other monitor records. (And, as mentioned in another append, the default SAMPLE CONFIG size is often too small for so many devices and has to be made larger.) But there's one general rule that has not yet been mentioned in this thread: don't let the MONDCSS overlay the storage of the virtual machine that is doing the data collecting, in this case PerfKit, or it will not be able to use it. While your MONDCSS looks VERY large to me, I'm admittedly out of date as far as current I/O configurations are concerned, and you apparently ended up with it for a good reason, after a trial and error phase with smaller sizes. Can you tell me the number of I/O devices that your VM sees and is collecting data for? Eginhard - Original Message - From: Bill Munson william.mun...@bbh.com That does not look like it is large enough. here is my definition MONDCSS CPDCSS N/A08000 0 SC R It can work for a while but if the segment is not large enough it will soon fail.
Re: [?? Probable Spam] Re: Perfkit SAMPLE CONFIG size too small
Not sure if my previous note on this was distributed. There is NO problem with having an oversized MONDCSS. A small one however stops you from collecting data. Our default is now 64mb. RPN01 wrote: Our MONDCSS grew, perhaps too large, while fighting this type message a long time ago. Once the problem was resolved, we didn't attempt to back off the changes we'd made, and the large size doesn't seem to hurt anything at the moment. I know that ultimately, making the segment larger was not the answer to the problem at the time, either. Also, Mr. Nunsford says hello.
Re: z/OS and UFT(D)
Check http://velocitysoftware.com/customer/tips/MULTSYST.html; (customer area). does that help? Mark Wheeler wrote: Is anyone sending files from z/OS to z/VM via z/VM's UFTD server? Would sure be nice, since we don't have RSCS. I checked doc for the TSO TRANSMIT command (most logical place) but (not surprisingly) nothing about UFT there. Best regards, Mark Wheeler UnitedHealth Group Your E-mail and More On-the-Go. Get Windows Live Hotmail Free. Sign up now. http://clk.atdmt.com/GBL/go/201469229/direct/01/
Re: CRYPTO card query
Martin, did anything show up on the crypto column on esausr4? Magat, Martin wrote: Hi May I know how to display the crypto card is being used on the zVM level by the guest? i.e. using CRYPTO APVIRT .. Query CRYPTO APQ does not show it ... Thanks
Re: zVM CPU allocation
As Rob and Alan have less blatantly stated, Linux CPU numbers are bogus in a virtual environment. The CPU reporting problem has been corrected in ESALPS (now zVPS) for those that want to use linux numbers for anything useful. If you don't have a mechanism to correlate the linux cpu numbers to reality, then one would not want to take the virtual reality too serious David Dean wrote: On Thu, 12 Nov 2009 13:01:52 -0600, Alan Ackerman alan.acker...@bankofamerica.com wrote: Where are you getting the 99% number? Which version of Linux are you running? Older versions of Linux were fooled by the CPU being taken away and reported high values of CPU utilization in 'top' and elsewhere. If all the numbers are from PerfKit, they don't make sense. You cannot have 3 users running at 99% and have the 3 IFLs running at 10-15% each. Are you sure you are looking at the same time interval? If you are looking at both Linxu and VM tools, they may not match up. Alan Ackerman Alan (dot) Ackerman (at) Bank of America (dot) com On Thu, 12 Nov 2009 13:17:26 -0500, Dean, David (I/S) david_d...@bcbst.com wrote: OK, this is for all you guys that understand the magic tunnel in which CPU processing (usage) flows from an IFL to a zLinux server. We have a z10 running zVM 5.4 in a dedicated LPAR with 3 dedicated IFL's. We have approximately 30 zLinux servers. Using IBM PerfKit, I list all of the individual USER / zLinux CPU usages, each of which generally run in the 1 to 5 % range. These servers do of course peak higher but on average they are pretty low. We then take a look at the 3 IFL's and see usage of maybe 5% on each. Now, let's say we have a USER / server or two or three go berserk and peak CPU at 99 % for an extended period of time. We then look at the IFL CPU usages and all three have climbed to maybe 10 to 15% each. How did the CPU get allocated? Is it always spread evenly across the IFL's, is that a setting? Why, if there were 3 USER / servers running at 99%, was more CPU not allocated from the 3 IFL's? Why did the IFL's decide to allocate X amount and go no further. In the USER DIRECTORY I allocate storage but not CPU. David M. Dean Information Systems BlueCross BlueShield Tennnessee - Please see the following link for the BlueCross BlueShield of Tennessee E- mail disclaimer: http://www.bcbst.com/email_disclaimer.shtm
Re: zVM CPU allocation
Mike, Rob probably left you out of the loop when he showed the new bogusness of the data with Sles10/11. It's more fun showing customers... Michael MacIsaac wrote: “less bogus”? More meaningful? Mike MacIsaac mike...@us.ibm.com (845) 433-7061
Re: VMRMSVM - z/VM Resource Manager
Terry, I would ask if you use it AND HAVE VALIDATED RESULTS? I've seen several sites install it during early days when there was no contention. So no problems means it is working? But when there is contention, the question is does it help when there is contention, or does it force servers to abend Martin, Terry R. (CMS/CTR) (CTR) wrote: Hi, I am looking at implementing VMRM. I was wondering if you use it and if it is working as advertised? I want to mainly use it for managing the priority of my different workloads running in z/Linux. I am familiar with the goal concept from WLM on the z/OS side so I understand the principle behind it but I just wanted to know from those who use it how it is working. Also any specifics on setting it up in terms of what to watch out for etc…. //Thank You,// //Terry Martin// //Lockheed Martin - Information Technology// //z/OS z/VM Systems - Performance and Tuning// //Cell - 443 632-4191// //Work - 410 786-0386// //terry.mar...@cms.hhs.gov mailto:terry.mar...@cms.hhs.gov// //WFH on Tuesdays and Fridays//
Re: Sending an SNMPTRAP alert from Velocity to NETCOOL/OMNIBUS console
I'm interested too. I've not been able to get omnibus doc so would be great if someone could provide doc on what is needed. We (Velocity Software/ESALPS/zVPS) provide a lot of data and a lot of alerts. Sending alerts somewhere that does not acknowledge doesn't do anyone any good, and we are probably the best to provide snmp trap prob rules. So, if somebody can help, it is time we (Velocity) provide assistance in this area Martin, Terry R. (CMS/CTR) (CTR) wrote: Hi I am trying to send an ALERT captured in Velocity ESAMON over to our NETCOOL/OMNIBUS console. I have the Velocity piece set up and we can see that the ALERT gets to the OMNIBUS console. See logs below. However nothing is being done with it because the OMNIBUS guy tells me that there needs to be a SNMP TRAP PROBE RULE defined. This is where it gets a little foggy for me. What should this rule look like on the OMNIBUS side to pick up my ALERT and have an email sent out based on the alert? I am also a little confused between a MIB and an OID in some of the documentation I read they seem to be used interchangeable. Anyway here is the SNMP TRAPDEST file that is defined in ESATCP: * THIS FILE IS THE LIST OF SNMP TRAP DESTINATIONS * FORMAT IS IP ADDRESS, COMMUNITY NAME, AND OPT OID 158.xx.xxx.xxx velocity 2B0601020102020102 Here is the log from the OMNIBUS console. I added ‘X’s to the IP address for privacy sake.: 10/15/09 14:37:12: Debug: 1 trap in queue 10/15/09 14:37:12: Debug: V1 trap received 10/15/09 14:37:12: Information: Number of items in the trap queue is 0 10/15/09 14:37:14: Debug: 158.xx.xxx.226: gethostbyaddr: h_errno = 1 10/15/09 14:37:21: Debug: 158.xx.xxx.226: gethostbyaddr: h_errno = 1 10/15/09 14:37:21: Debug: [Event Processor] ReqId: 0 10/15/09 14:37:21: Debug: [Event Processor] enterprise: .1.3.6.1.4.1.15601 10/15/09 14:37:21: Debug: [Event Processor] generic-trap: 6 10/15/09 14:37:21: Debug: [Event Processor] specific-trap: 0 10/15/09 14:37:21: Debug: [Event Processor] UpTime: 845 10/15/09 14:37:21: Debug: [Event Processor] Uptime: 0:00:08.45 10/15/09 14:37:21: Debug: [Event Processor] community: velocity 10/15/09 14:37:21: Debug: [Event Processor] IPaddress: 158.xx.xxx.226 10/15/09 14:37:21: Debug: [Event Processor] PeerIPaddress: 158.xx.xxx.226 10/15/09 14:37:21: Debug: [Event Processor] ReceivedPort: 162 10/15/09 14:37:21: Debug: [Event Processor] ReceivedTime: 1255610232 10/15/09 14:37:21: Debug: [Event Processor] Protocol: UDP 10/15/09 14:37:21: Debug: [Event Processor] SNMP_Version: 1 10/15/09 14:37:21: Debug: [Event Processor] OID1: .1.3.6.1.2.1.2.2.1.2 10/15/09 14:37:21: Debug: [Event Processor] 1: PAGE SPACE IS 34.43% USED 10/15/09 14:37:21: Debug: [Event Processor] 1_raw: PAGE SPACE IS 34.43% USED 10/15/09 14:37:21: Debug: [Event Processor] 1_text: PAGE SPACE IS 34.43% USED 10/15/09 14:37:21: Debug: [Event Processor] 1_hex: 50 41 47 45 20 53 50 41 43 45 20 49 53 20 20 33 34 2e 34 33 25 20 55 53 45 44 20 10/15/09 14:37:21: Debug: [Event Processor] .1.3.6.1.2.1.2.2.1.2: PAGE SPACE IS 34.43% USED 10/15/09 14:37:21: Debug: [Event Processor] Node: 158.xx.xxx.226 10/15/09 14:37:21: Debug: [Event Processor] PeerAddress: 158.xx.xxx.226 10/15/09 14:37:21: Debug: [Event Processor] EventCount: 1355 10/15/09 14:37:21: Debug: [Event Processor] Processing alert {0 remaining} 10/15/09 14:37:21: Debug: Entering... snmptrap.rules 10/15/09 14:37:21: Debug: Entering... FixMttrapdOids.include.snmptrap.rules 10/15/09 14:37:21: Debug: Leaving... FixMttrapdOids.include.snmptrap.rules 10/15/09 14:37:21: Debug: (snmptrap.rules) Event is an enterprise-specific trap. 10/15/09 14:37:21: Debug: (snmptrap.rules) Enterprise ID not found, checking ncotdc include files. 10/15/09 14:37:21: Debug: (snmptrap.rules) Enterprise ID not found in any include file. 10/15/09 14:37:21: Debug: Entering... CorrScore.include.snmptrap.rules 10/15/09 14:37:21: Debug: A value for '' doesn't exist in lookup table 'snmptrapCorrScore' - using default 10/15/09 14:37:21: Debug: Leaving... CorrScore.include.snmptrap.rules 10/15/09 14:37:21: Debug: Entering... PreClass.include.snmptrap.rules 10/15/09 14:37:21: Debug: A value for '' doesn't exist in lookup table 'snmptrapPreClass' - using default 10/15/09 14:37:21: Debug: Leaving... PreClass.include.snmptrap.rules 10/15/09 14:37:21: Debug: Entering... omnibus36.include.compat.rules 10/15/09 14:37:21: Debug: (omnibus36.include.compat.rules) $OPTION_TypeFieldUsage NOT set to 3.6. 10/15/09 14:37:21: Debug: Leaving... omnibus36.include.compat.rules 10/15/09 14:37:21: Debug: Entering... AdvCorr36.include.compat.rules 10/15/09 14:37:21: Debug: Leaving... AdvCorr36.include.compat.rules 10/15/09 14:37:21: Debug: Leaving... snmptrap.rules 10/15/09
Re: Reaction of z/VM when losing a page dataset
The problem with putting all your paging volumes on one DS-8000 then becomes performance. Your paging devices will now be sharing physical disks. When you write pages out to multiple page devices, then all of the I/O is then going to the same set of hardware. From a performance perspective, you want to alternate page devices across ranks of physical devices. It is rare that you will lose the DS-8000. If you do, there are a lot of other things to worry about. Would be less optimal to have bad paging performance because of worrying about an outage that is very unlikely to happen. Florian wrote: Thank you, This is what I expected. I fear also that the chance is doubled to crash the system when the PAGING volumes are spread on both DS-8000. So I will correct this soon.
Re: MIPS CAPPING z/VM Equivalent
If you have the ability to capture CPU seconds consumed (a good performance monitor does this), then it is an easy calculation assuming you know the power of your IFL: (CPU Seconds / time) * MIPSRating Paul, Thomas wrote: Hi We are in the process of migrating Websphere applications from z/OS to z/Linux under VM. I have questions, if you can answer these questions, it'll be greatly apprciated. 1. Under z/OS environment MIPS capping was implemented. Is there equivalent in z/VM - I know of Share (abs rel), setting VM tuning besides all of that and VMRM. 2. Is there a MIPS count vs cpu utilization conversion table? 3. Can anyone share comparative statistics z/OS vs. z/VM? Thank you in anticipation. Regards Tom 1-508-395-9374
Re: Linux Sizing z/VM Customization
The Q3 issue is likely a websphere polling issue that Rob is working with WAS development on. The solution to that currently leads back to using CMM1 as well. Will call. Paul, Thomas wrote: Hi Barton, Thank you. I would like to explore that possibility. Could you give me a call please. 1-508-395-9374. Another problem, I'm facing is all our Linux guests are in Q3 (Rel. share 100). I like to change that since they need not be in Q3 all the time in PS. We are running Sles 9. Any thought on timing in Linux to make this work - SRM planned changes are Storbuf 300 300 300 Ldubuf (don't remember the numbers). Regards Tom -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Barton Robinson Sent: Thursday, August 27, 2009 6:03 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Linux Sizing z/VM Customization CMM-1 is almost always appropriate (SLES9, SLES10, RHEL4, RHEL5). CMMA is NOT. CMM1 is recommended, CMMA is not. Mark is right on the numbers. Paul, Thomas wrote: Hi Mark, CMM is not applicable here because of the S/W H/W. 1M is nothing to brag about. Thank you for the input. Thanks Tom -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Mark Post Sent: Thursday, August 27, 2009 4:11 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Linux Sizing z/VM Customization On 8/27/2009 at 10:26 AM, Paul, Thomas thomas.p...@iso.com wrote: -snip- 2. Linux Sizing - All Linux guests are independent - in other words, they all have their own Kernel, etc. The idea is to build an NSS and make it like a CMS user. So, if I accomplish that what would be the size of z/Linux guest under z/VM with just z/Linux running. And, second if I do build DCSS for Websphere binary, how much storage would I be able to save? Currently, most of them running at 1.2G 1.5G. If by build an NSS you mean having the kernel in an NSS, that will save you about 1MB per guest that uses it. Not a whole lot. According to Barton Robinson of Velocity Software, you get the biggest real storage savings by using CMM and xip2fs. (If I'm remembering wrong, I know Barton will correct this.) CMM is the easiest to implement, and doesn't really require any effort to maintain. Setting up xip2fs is not terribly easy to set up (I'm working on getting that changed) and not easy to maintain. Still, if you're really constrained, it may be worth the effort. For some insight into that process, look at the presentation on it at http://linuxvm.org/Present/ Mark Post This email is intended for the recipient only. If you are not the intended recipient please disregard, and do not use the information for any purpose. This email is intended for the recipient only. If you are not the intended recipient please disregard, and do not use the information for any purpose.
Re: Linux Sizing z/VM Customization
I am NOT inferring using VMRM. From what I can tell, VMRM has no feed back mechanism, and experience at several installations are that because of this, VMRM has taken so much storage away from servers that the server or application dies. I would HIGHLY recommend against using it. VMRM also has an added design flaw that if you've seen my Linux storage presentation, you would understand. When paging gets bad, VMRM will work very hard to increase the paging load. Using zVPS (VelocitySoftware.com/zVPS.HTML (new name for ESALPS), we do have feedback mechanisms, we know what storage is in use INSIDE the linux server, and we can quickly determine impact of CMM1 commands on the server - and can detect when the workload changes. Thus using the CMM1 commands takes knowledge of how it really works (see my storage presentation), takes feedback, and takes the ability to react. Some of this right now needs to be manual - I hope to get something out in very near term to automate this function using zVPS. And re cpuplugd, sorry, really don't like that, it really results in changing scheduler settings dynamically - which means that performance settings need much better understanding with respect to workload requirements. Sterling James wrote: Barton, When you say using CMM1 are you inferring using the CMM-VMRM or another method like cpuplugd? Thanks *Barton Robinson bar...@vm1.velocity-software.com* Sent by: The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU 08/28/2009 12:03 PM Please respond to The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU To IBMVM@LISTSERV.UARK.EDU cc Subject Re: Linux Sizing z/VM Customization The Q3 issue is likely a websphere polling issue that Rob is working with WAS development on. The solution to that currently leads back to using CMM1 as well. Will call. Please consider the environment before printing this email and any attachments. * This e-mail and any attachments are intended only for the individual or company to which it is addressed and may contain information which is privileged, confidential and prohibited from disclosure or unauthorized use under applicable law. If you are not the intended recipient of this e-mail, you are hereby notified that any use, dissemination, or copying of this e-mail or the information contained in this e-mail is strictly prohibited by the sender. If you have received this transmission in error, please return the material received to the sender and delete all copies from your system. *
Re: Linux Sizing z/VM Customization
CMM-1 is almost always appropriate (SLES9, SLES10, RHEL4, RHEL5). CMMA is NOT. CMM1 is recommended, CMMA is not. Mark is right on the numbers. Paul, Thomas wrote: Hi Mark, CMM is not applicable here because of the S/W H/W. 1M is nothing to brag about. Thank you for the input. Thanks Tom -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Mark Post Sent: Thursday, August 27, 2009 4:11 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Linux Sizing z/VM Customization On 8/27/2009 at 10:26 AM, Paul, Thomas thomas.p...@iso.com wrote: -snip- 2. Linux Sizing - All Linux guests are independent - in other words, they all have their own Kernel, etc. The idea is to build an NSS and make it like a CMS user. So, if I accomplish that what would be the size of z/Linux guest under z/VM with just z/Linux running. And, second if I do build DCSS for Websphere binary, how much storage would I be able to save? Currently, most of them running at 1.2G 1.5G. If by build an NSS you mean having the kernel in an NSS, that will save you about 1MB per guest that uses it. Not a whole lot. According to Barton Robinson of Velocity Software, you get the biggest real storage savings by using CMM and xip2fs. (If I'm remembering wrong, I know Barton will correct this.) CMM is the easiest to implement, and doesn't really require any effort to maintain. Setting up xip2fs is not terribly easy to set up (I'm working on getting that changed) and not easy to maintain. Still, if you're really constrained, it may be worth the effort. For some insight into that process, look at the presentation on it at http://linuxvm.org/Present/ Mark Post This email is intended for the recipient only. If you are not the intended recipient please disregard, and do not use the information for any purpose.
Re: PAV's
a good performance too would tell you this. Dean, David (I/S) wrote: Is there a way by looking at the device numbers below to tell which is a PAV? I know how to go back and Q PAV the answer but is there any way to tell by looking, a way the numbers are generated that would tell me? 0 * * * Top of File * * * 1 B01A L7601A 2 B0D4 L7601A 3 C31E L7602A 4 C395 L7602A 5 C41E L7603A 6 C4F2 L7603A 7 * * * End of File * * * David Dean Information Systems *bcbstauthorized* - Please see the following link for the BlueCross BlueShield of Tennessee E-mail disclaimer: http://www.bcbst.com/email_disclaimer.shtm
Re: MP effect on z/VM Linux hosting
First, IBM has never measured this for z/VM, their resources devoted to things like LSPR for VM are long gone. So you are either getting a sale's person's guess, or a z/OS guess and I guess they guess different? Amdahl's Law of Multiprogramming from a software perspective states that a system's thruput (in regards to adding processors) is governed by the amount of code that is single threaded. z/VM has done a very good job of getting code off of the master processor, and reducing or eliminating locking, and Linux workloads unless very storage constrained with lots of emergency scans don't abuse the master. With z/VM's affinity mechanism with the Processor Local Dispatch Vector where one virtual machine block is associated with a processor, the processor cache is also better utilized. The association dies if there are idle processors looking for work, in which case losing a cache is not important because there are extra processors sitting idle looking for work. And there is the other option of multiple lpars with dedicated IFLs, to reduce impact of dropping caches. Best guess - we want ibm to lower the prices so agree with ibm that 359 is the right number, in planning, MY guess is that it would be VERY MUCH higher. But the reality is that with the current storage problems that IBM software group is giving us with their polling in WAS, DB2, Domino and SAP are going to keep this a non-problem for a while. (Though DB2 has very recently made very good progress toward becoming virtual friendly). Currently my guess is that until ibm fixes their software you can't/won't buy enough real storage to support 60 z10/xx processors. Marcy Cortes wrote: I'm getting conflicting answers from IBM. How does VM scale with regards to multiprocessors? In the z/OS world the 1st z/10 engine is like 900 something MIPS and the 60th on the box is something like 359. Does z/VM hosting Linux suffer the same fate? (z/OS per engine pricing actually goes down to compensate for this, z/VM's does not). Marcy This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose, or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.
Re: SWAPGEN
make sure that your oracle SGA fits into your page cache. If it doesn't, that will make you swap. Your ORACLE DBA ABSOLUTELY must be in agreement on your configuration Martin, Terry R. (LOCKHEED MARTIN Performance Engineering/CTR) (CTR) wrote: Hi Barton, Thanks for the information. I will have two VDISKs for my z/Linux guests already. I have found that when a z/Linux guest at least with my workload starts to use SWAP it plows through it in no time so the extra real disk for swap was to slow this down so that I could react quickly to head of off running out of SWAP. In some cases with some of my Oracle workload by the time I received the alert that the second VDISK was being used it would have already been used up. My paging subsystem is pretty robust as it stands and I do very little paging so far. Thank You, Terry Martin Lockheed Martin - Information Technology z/OS z/VM Systems - Performance and Tuning Cell - 443 632-4191 Work - 410 786-0386 terry.mar...@cms.hhs.gov -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Barton Robinson Sent: Wednesday, April 29, 2009 6:42 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: SWAPGEN giving real disks to swap is a real waste of resource. It is much better to take the extra disk resource that you allocate but never want to use, and assign it to z/VM paging to enhance your paging subsystem. Then define two vdisks for swap, prioritize them, and set an alert when the 2nd disk is being used. Martin, Terry R. (LOCKHEED MARTIN Performance Engineering/CTR) (CTR) wrote: Hi I am using SWAPGEN to define by z/Linux VDISKS I also want to define a real disk for swap. My question is can I use SWAPGEN to define a swap on real DASD? If you have an example of the control card syntax to accomplish this that would be great? //Thank You,// //Terry Martin// //Lockheed Martin - Information Technology// //z/OS z/VM Systems - Performance and Tuning// //Cell - 443 632-4191// //Work - 410 786-0386// //terry.ma...@cms.hhs.gov mailto:terry.ma...@cms.hhs.gov//
Re: SWAPGEN
giving real disks to swap is a real waste of resource. It is much better to take the extra disk resource that you allocate but never want to use, and assign it to z/VM paging to enhance your paging subsystem. Then define two vdisks for swap, prioritize them, and set an alert when the 2nd disk is being used. Martin, Terry R. (LOCKHEED MARTIN Performance Engineering/CTR) (CTR) wrote: Hi I am using SWAPGEN to define by z/Linux VDISKS I also want to define a real disk for swap. My question is can I use SWAPGEN to define a swap on real DASD? If you have an example of the control card syntax to accomplish this that would be great? //Thank You,// //Terry Martin// //Lockheed Martin - Information Technology// //z/OS z/VM Systems - Performance and Tuning// //Cell - 443 632-4191// //Work - 410 786-0386// //terry.ma...@cms.hhs.gov mailto:terry.ma...@cms.hhs.gov//
Re: FTP Server very slow
In looking at many cases like this, it was NEVER the network or TCPIP, but something else on z/VM, sometimes Escon channels, sometime device contention, sometimes cache in the storage controller, sometimes CPU contention. I would recommend a good z/VM performance monitor Salecky, Zenko J wrote: I have copied netstat pool output from both the client and server side. The valus pretty much stay the same throughout the file transfer time. I also looked at the TCPIP console and di not see any messages relating to buffer shortages. ftp Server side netstat pool VM TCP/IP Netstat Level 540 TCP/IP Server Name: TCPIP TCPIP Free pool status: ObjectNo. allocNo. freeLo-water Permit size === === ACB102410161009 102 CCB 154 143 14015 Dat buf 160 148 14616 Sm dat buf 12 10 7 1 Tiny dat buf 10 10 9 1 Env 750 750 74275 Lrg env 50 49 42 5 RCB 51 51 51 5 SCB 264 257 25726 SKCB256 256 25425 TCB 256 251 25125 UCB 102 102 10010 Add Xlate 151315081502 5 NCB150115011501 5 IP Route609 607 607 6 IPv6 Route 605 605 605 6 Segment ACK512051205112 512 FPSP total locked pages: 334, Unused locked pages: 55 FPSP allocation threshold: 23984, Low-water mark: 0 TCPIP machine size: 128M, Pools: 5013K, Avail: 105040K, Max block: 85584K Ready; T=0.01/0.01 08:38:33 ftp client side netstat pool VM TCP/IP Netstat Level 540 TCP/IP Server Name: TCPIP TCPIP Free pool status: ObjectNo. allocNo. freeLo-water Permit size === === ACB102410161003 102 CCB 154 136 12515 Dat buf 160 154 14616 Sm dat buf 12 9 7 1 Tiny dat buf 10 9 1 1 Env 750 750 74175 Lrg env 50 49 42 5 RCB 51 51 43 5 SCB 264 256 24826 SKCB256 255 24525 TCB 256 249 24025 UCB 102 102 9210 Add Xlate 151315081508 5 NCB150115011501 5 IP Route609 607 607 6 IPv6 Route 605 605 605 6 Segment ACK512051205112 512 FPSP total locked pages: 344, Unused locked pages: 65 FPSP allocation threshold: 2220, Low-water mark: 0 TCPIP machine size: 32M, Pools: 5013K, Avail: 8304K, Max block: 8296K Ready; T=0.01/0.01 08:42:32 and transfer is still very slow 10639661 bytes transferred. 10796013 bytes transferred. 10952702 bytes transferred. 250 Transfer completed successfully. 11030147 bytes transferred in 655.865 seconds. Transfer rate 16.82 Kbytes/sec. Command: If I
Re: Monitor versus accounting data
Ok, let's go through this one at a time. See inserted comments. Alan Ackerman wrote: Creating new thread. 1. The folks that receive the data at my shop are z/OS folks. Historicall the capture ratio of MVS was really poor. The notion was that you should use SMF data and never RMF data. I don't know if z/OS has cleaned up its act or not. But I have heard the same thing from VM folks. (I've said it myself.) As Barton says, the capture ratio in VM has always been quite high, due t the way the data is captured in the VMDBK. However, Barton computes this (I think) by comparing different record types in the monitor data, not by comparing monitor to accounting data. There is system overhead, but it is captured in the SYSTEM VMDBK block. Accounting data and monitor data are using the same data, so they should get the same results. Of course, some time gets charged to the wrong user for example between the time an interrupt comes in and the new user is identified. But it shows up the same in the monitor and the accounting data. (User CPU time is more reproducible than total CPU time, for this reason.) Is some time gets charged to the wrong user a validated and relevant issue? I've not seen any overhead issues in accounting or monitor data in MANY years. 2. Monitor sample data is taken at one minute samples. It used to be that data for users that logged in or off between samples was dropped for the partial minutes. Is this still true? Was it ever true? Or is it urban folklore? Transaction records are cut at logon/logoff, that is how we get 100.00% capture ratio. Nothing is lost. 3. On our systems, we sometimes see messages from CP that say the monitor data has been thrown away because the user connected to *MONITOR did not respond in time. This happens when the system is overloaded, either in CP or storage. So we lose some minutes of monitor data, but not, I think, accounting data. Often you can fix this by increasing the segment sizes or give MONWRITE/ESAWRITE a bigger SHARE. Not always, though. In some cases the monitor segments get paged out. (We reported it to Velocity, who said it was a CP problem.) I think IBM could do things to make collection of monitor data more reliable in the extreme cases. Unfortunately, I'm not responsible for this and it is only performance data. I think this can be dealt with, but it does take diligence and wor to keep your monitor data accurate. You don't have to do this work for accounting data. I think IBM could do things to make collection of monitor data more easy. This still does happen occasionally when systems are thrashing so much that everything stops. At this point, accounting is probably lower priority. Capacity planning and performance tuning do need to be employed in this platform. IBM could stop the DCSS from being paged out when the system starts to thrash. 4. On our systems, we switch files (I think hourly) to keep them from getting too big. We lose a minute or two of data each time. ESALPS does not lose data each hour. Capture ratio is 100% 5. The default for ESAWRITE is to collect User history records only for userids using more than 0.5% CPU. So when we go back to process CPU utilization for users, we get smaller totals for monitor than from accounting data. I assume this could be fixed by setting the threshold to zero. I don't know which of these, if any, affect the ESALPS data collection that Barton mentioned. We have tested ESALPS, but are not yet licensed. The default for ESAWRITE is 100% capture ratio. ALL USER DATA is captured and retained for capacity planning and accounting. the thresholds only apply to current performance data. This has been the case for 20 years. I'll repeat, capture ratio for user data is ALWAYS 100.00%. You can't look at the interval data collected for performance and use it for accounting. The summary data for each hour is 100% and is what one would use for accounting and capacity planning. Alan Ackerman Alan (dot) Ackerman (at) Bank of America (dot) com
Re: Monitor versus accounting data
logoff record happens when the virtual machine leaves the building, forced or voluntarily. Lionel B. Dyck wrote: Barton said: Transaction records are cut at logon/logoff, that is how we get 100.00% capture ratio. Nothing is lost. Question: Is there a interval record to capture the data should the system go south unexpectedly and/or the user is forced off (or is the logoff record cut at force) ? *Lionel B. Dyck, Consultant/Specialist * Enterprise Platform Services, Mainframe Engineering KP-IT Enterprise Engineering 925-926-5332 (8-473-5332) | E-Mail: _lionel.b.d...@kp.org_ mailto:lionel.b.d...@kp.org AIM: lbdyck *|* Yahoo IM: lbdyck / Kaiser Service Credo: Our cause is health. Our passion is service. We’re here to make lives better.” / * “Never attribute to malice what can be caused by miscommunication.” * * NOTICE TO RECIPIENT: *If you are not the intended recipient of this e-mail, you are prohibited from sharing, copying, or otherwise using or disclosing its contents. If you have received this e-mail in error, please notify the sender immediately by reply e-mail and permanently delete this e-mail and any attachments without reading, forwarding or saving them. Thank you.
Re: Monitor versus accounting data
If the LPAR leaves the building, then the data that is captured is valid for the previous hour. The data for the current hour contains data each minute for the top 10 consumers - so for the current partial hour, capture ratio might only be 95%, and would have to be recovered from the interval/performance data. Lionel B. Dyck wrote: But not if the lpar (z/vm) leaves the building without saying goodbye *Lionel B. Dyck, Consultant/Specialist * Enterprise Platform Services, Mainframe Engineering KP-IT Enterprise Engineering 925-926-5332 (8-473-5332) | E-Mail: _lionel.b.d...@kp.org_ mailto:lionel.b.d...@kp.org AIM: lbdyck *|* Yahoo IM: lbdyck / Kaiser Service Credo: Our cause is health. Our passion is service. We’re here to make lives better.” / * “Never attribute to malice what can be caused by miscommunication.” * * NOTICE TO RECIPIENT: *If you are not the intended recipient of this e-mail, you are prohibited from sharing, copying, or otherwise using or disclosing its contents. If you have received this e-mail in error, please notify the sender immediately by reply e-mail and permanently delete this e-mail and any attachments without reading, forwarding or saving them. Thank you. From: Barton Robinson bar...@vm1.velocity-software.com To: IBMVM@LISTSERV.UARK.EDU Date: 04/03/2009 08:44 AM Subject:Re: Monitor versus accounting data Sent by:The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU logoff record happens when the virtual machine leaves the building, forced or voluntarily. Lionel B. Dyck wrote: Barton said: Transaction records are cut at logon/logoff, that is how we get 100.00% capture ratio. Nothing is lost. Question: Is there a interval record to capture the data should the system go south unexpectedly and/or the user is forced off (or is the logoff record cut at force) ? *Lionel B. Dyck, Consultant/Specialist * Enterprise Platform Services, Mainframe Engineering KP-IT Enterprise Engineering 925-926-5332 (8-473-5332) | E-Mail: _lionel.b.d...@kp.org_ mailto:lionel.b.d...@kp.org AIM: lbdyck *|* Yahoo IM: lbdyck / Kaiser Service Credo: Our cause is health. Our passion is service. We’re here to make lives better.” / * “Never attribute to malice what can be caused by miscommunication.” * * NOTICE TO RECIPIENT: *If you are not the intended recipient of this e-mail, you are prohibited from sharing, copying, or otherwise using or disclosing its contents. If you have received this e-mail in error, please notify the sender immediately by reply e-mail and permanently delete this e-mail and any attachments without reading, forwarding or saving them. Thank you.
Re: Guest Billing
ESALPS provides the data for accounting for the Linux process level, linux application level, linux user level, and of course at the virtual machine level. How to do this I thought was on our website, will put it there today. It involves a very simple process usually as part of the night time operations, extracts the data from the performance database, which contains the information you need for accounting. No problem even using the z/VM accounting codes. On Thu, Apr 2, 2009 at 8:15 AM, =?iso-8859-1?Q?Greg_Dyrda?= gregory.l.dy...@us.hsbc.com mailto:gregory.l.dy...@us.hsbc.com wrote: We currently bill for Linux on a per guest basis. I'm wondering what approach others are taking. Specifically, I'm wondering if it is possible to bill at the process level and if anyone else is billing that way.
Re: Guest Billing
Alan, ESALPS correlates the linux process data and the z/vm data, allowing chargeback to be done correctly at the process level. Other products have not announced this capability as far as i know? So you would be correct for other methods of collecting process data. Alan Ackerman wrote: On Thu, 2 Apr 2009 09:15:57 -0500, =?iso-8859-1?Q?Greg_Dyrda?= gregory.l.dy...@us.hsbc.com wrote: We currently bill for Linux on a per guest basis. I'm wondering what approach others are taking. Specifically, I'm wondering if it is possible to bill at the process level and if anyone else is billing that way. = === I think a Linux expert would have to answer this, but I don't think it is possible to charge-back accurately on a per-process basis when running under VM. The actual CPU time is captured by VM (CP) -- but it knows nothing about processes. The CPU utilization numbers per process that Linux produces in, e.g., TOP, are just plain wrong, because Linux does no t notice when the processor has been taken away and given to another guests . Linux thinks elapsed time while process is running = CPU time -- but of course it does not. I do remember information at SHARE about revisions to Linux to allow it t o know what the real processor utilization is (by issuing the appropriate DIAG instruction), but I don't know if that ever made it into the Linux shipped by Red Hat or SuSE/Novell. That’s why I said a Linux under VM expert would need to answer this. Right now we are working on moving VM accounting data to MICS on z/OS to do charge-backs, and that is by virtual machine, not by process. You could use the per-process CPU times in Linux to prorate the VM measured CPU time, but I doubt that would be accurate. Some processes are much more likely to have the CPU stolen than others. (Those that are CPU- bound instead of I/O-bound, for example.) Alan Ackerman Alan (dot) Ackerman (at) Bank of America (dot) com
Re: Guest Billing
ESALPS uses MONITOR data, that has 100.00% capture ratio, accurate to the micro second for z/VM. For Linux data, we capture the process table every minute to correlate to the vm monitor interval. I believe the capture ratio obtained by ESALPS from monitor data is higher that what you get from accounting data, but i do measure capture ratio (and have for 20 years) and it is very very good for monitor data. Alan Ackerman wrote: The ESAPLPS code that Barton mentioned is doing prorating, right? How accurate is that? Are you using accounting data or performance data? I have a long-standing bias against using performance data for accounting . I can go into detail, if anyone needs to know, but first we have to see which data Barton is using. Alan Ackerman Alan (dot) Ackerman (at) Bank of America (dot) com
Re: Guest Billing
Oh, and yes, I too would charge based on resident storage (and yes we have that data available). Charging on resident storage would make things like current WAS polling show up as expensive as it really is - it ensures Linux doesn't easily page out. And then it would be cheaper for the chargee to use smaller virtual machines, focusing your users on minimizing the resource requirements. Scott Rohling wrote: Just a quick note that one thing I see missing from most billing schemes is memory usage. How much memory is assigned to the guest can have more of an impact on the system than CPU, depending on the environment. It's also easier to monitor as the virtual machine guest size stays fairly static. So I would include the memory as part of the billing scheme.. charge a fair amount per 1G of memory to discourage apps from grabbing large chunks of it 'just because'. Scott
Re: Guest Billing
exactly. installations wanting to charge at the process level or anything to do with inside linux using valid data will use ESALPS. Traditional vm sites not interested in what is inside linux can use a multiple of data sources available for 30 years or so. this thread was about linux process level, for which there is NO valid accounting data, only valid performance data that is used by many for accounting. Alan Ackerman wrote: On Thu, 2 Apr 2009 10:09:08 -0800, Barton Robinson bar...@vm1.velocity- Software.com wrote: Alan, ESALPS correlates the linux process data and the z/vm data, allowing chargeback to be done correctly at the process level. Other products have not announced this capability as far as i know? So you would be correct for other methods of collecting process data. Does it use monitor data or accounting data? People at my shop would not like it using monitor data. Other shops can decide for themselves. Alan Ackerman Alan (dot) Ackerman (at) Bank of America (dot) com
Re: Guest Billing
Because I get asked often about accounting and charge back for Linux processes, z/VM virtual machines, and Linux applications, I've put up more material explaining what data is available, and how to get that data. The web page is at http://www.VelocitySoftware.com/account.html; =?iso-8859-1?Q?Greg_Dyrda?= wrote: We currently bill for Linux on a per guest basis. I'm wondering what approach others are taking. Specifically, I'm wondering if it is possibl e to bill at the process level and if anyone else is billing that way.
zLinux / Oracle Conference
If anyone is interested in Oracle on zLinux, this conference: http://www.zseriesoraclesig.org/; would be of great interest. It will be a pretty intense week, and for the price is one of the best price performers in the conference calender.
Re: Paging
In ESALPS, ESAUSPG shows by user. ESAUCD2 shows if you have extra cache or buffer. If you can send your reports, we will analyze it at no charge. Martin, Terry R. (CMS/CTR) (CTR) wrote: Hi I seem to be doing a lot of paging currently on my z/VM 5.3 system I am running multiple Linux guests including a large Oracle guest (40 GB memory size). How can I find out 1) who is doing the majority of the paging and along with that 2) I believe that some of the paging slots are old data in other words the pages are not going away after a task is complete how can I research this. The Linux guests have not been recycled but I thought if they had allocated the slots that after a task within the Linux guest completed that the slots would be reclaimed. Any thoughts on all of this would be appreciated. //Thank You,// //Terry Martin// //Lockheed Martin - Information Technology// //z/OS z/VM Systems - Performance and Tuning// //Cell - 443 632-4191// //Work - 410 786-0386// //terry.ma...@cms.hhs.gov mailto:terry.ma...@cms.hhs.gov//
Re: Correcting Statements From Marketing
If you build a response time model for processors, - AND you have a target response time not to be exceeded, it is easy to show that 1 processor responds worse at 80%, than two at 80%. Equivalent response time is expected when the two processors are at 90%. So the source of the question is really batch mentality vs the response time mentality. MP effect comes from the batch mentality where thruput was the only measure. The batch mentality will always challenge this, response time mentality should understand If you care about response time in the Linux/zVm world, you don't run at 100% most of the time. So the only time the MP Effect question is relevant is when both processors are running at 100%, which makes the question not relevant on IFLs. From an accounting perspective, I guess you could use the z/OS numbers, which would likely under-charge the Linux user for CPU consumed, since using those numbers a CPU second consumed is not charged as a full CPU second. Schuh, Richard wrote: I would expect that some would challenge your conclusion based on the idea that the MP effect does not even appear unless you are running at or near capacity. If I have two cpus or IFLs and 1.1 cpu's worth of demand, will I notice the MP effect? Probably not. I probably will see a better service level than when I was trying to service the same demand with only 1 cpu. The question is, if n tasks causes a single engine to run at 100%, will 2 engines be able to service 2n tasks as well as 1 serviced n? I think that under normal circumstances, the answer is that the 2 engine machine will only be able to service somewhat less than 2n. Regards, Richard Schuh -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Barton Robinson Sent: Tuesday, February 03, 2009 10:57 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Correcting Statements From Marketing Ok here's some heresy that I've presented to IBM and maybe was communicated to their sales folks. From a capacity planning and service level perspective, adding a CPU gives you MORE than 100%, not less than. Really, BUT ONLY if you actually care about service levels. From a service level perspective, i know that i can provide on ONE IFL a given service at 80% CPU utilization. If I ADD an IFL, and more work of a similar nature, I now have TWO IFLs, and I know that I can provide that SAME service at 180% CPU Util. So, I went from ONE IFL, to TWO IFLs, and increased my target CPU utilization by 1.25 times. On z/OS if you just run at 100% all the time, and run batch to soak up cycles, then add a CPU and you don't get 100% of one CPU more work done. That is the only time MP factors should matter. And this heresy is why it is much easier to deal with installations running multiple IFLs, because the performance will be better at higher utilizations than single IFLs at lower utilizations. Adding a second IFL more than doubles your usable capacity. Adding a 3rd or 4th is less dramatic. From a historical perspective, we used to have the MASTER PROCESSOR effect where adding a CPU added much less capacity. Installations today do not see this impact. Schuh, Richard wrote: This got no response when posted under a different topic: Yikes, We have someone from IBM Marketing now making the statement, I have confirmed...no MP factor with IFLs. That is the entire statement, all of the dots included. I did not replace anything with ellipses. Somehow, that does not ring true. I mentioned that the rating of an IFL is the same as that of an ordinary CPU and someone went to marketing for the real answer. Perhaps they should have said, No different MP factor for IFLs than for regular CPUs, they are the same in that regard. That would make more sense. Anyone from IBM care to comment - you will probably be quoted. I am not considered an authority on the topic, especially when I disagree with an interpretation of a statement made by IBM marketing. I need to disabuse someone of their notion because it will affect the capacity planning process. They do not seem to believe that running the same O/S on two systems, one with n standard CPUs and the other with the same number of IFLs will produce a result of equal MP effect. Barton, you are also invited to respond. At least one of the people on the other side of the fence will take your word for it. Regards, Richard Schuh
Re: Correcting Statements From Marketing
Ok here's some heresy that I've presented to IBM and maybe was communicated to their sales folks. From a capacity planning and service level perspective, adding a CPU gives you MORE than 100%, not less than. Really, BUT ONLY if you actually care about service levels. From a service level perspective, i know that i can provide on ONE IFL a given service at 80% CPU utilization. If I ADD an IFL, and more work of a similar nature, I now have TWO IFLs, and I know that I can provide that SAME service at 180% CPU Util. So, I went from ONE IFL, to TWO IFLs, and increased my target CPU utilization by 1.25 times. On z/OS if you just run at 100% all the time, and run batch to soak up cycles, then add a CPU and you don't get 100% of one CPU more work done. That is the only time MP factors should matter. And this heresy is why it is much easier to deal with installations running multiple IFLs, because the performance will be better at higher utilizations than single IFLs at lower utilizations. Adding a second IFL more than doubles your usable capacity. Adding a 3rd or 4th is less dramatic. From a historical perspective, we used to have the MASTER PROCESSOR effect where adding a CPU added much less capacity. Installations today do not see this impact. Schuh, Richard wrote: This got no response when posted under a different topic: Yikes, We have someone from IBM Marketing now making the statement, I have confirmed...no MP factor with IFLs. That is the entire statement, all of the dots included. I did not replace anything with ellipses. Somehow, that does not ring true. I mentioned that the rating of an IFL is the same as that of an ordinary CPU and someone went to marketing for the real answer. Perhaps they should have said, No different MP factor for IFLs than for regular CPUs, they are the same in that regard. That would make more sense. Anyone from IBM care to comment - you will probably be quoted. I am not considered an authority on the topic, especially when I disagree with an interpretation of a statement made by IBM marketing. I need to disabuse someone of their notion because it will affect the capacity planning process. They do not seem to believe that running the same O/S on two systems, one with n standard CPUs and the other with the same number of IFLs will produce a result of equal MP effect. Barton, you are also invited to respond. At least one of the people on the other side of the fence will take your word for it. Regards, Richard Schuh
Re: Linux Guest 'swapping'
The last time I looked at the cost of swap to vdisk, at 1,000 per second, used 10% of an 890 processor. It's very hard to constrain a system to swap this much, this was in the lab pushing limits not normally pushed. With z10 IFL significantly faster, swapping to vdisk would not be a significant cost. The largest performance problem facing us today is storage, as IBM Software Group has decided to put polling back into their applications (remember the hertz timer in Linux we eliminated in 2003? - it's back courtesy of IBM applications). With polling, the over commit ratio you can attain is now about 1.5 - so reducing Linux storage sizes and causing some swap means more Linux servers per installed storage. Robert J Brenneman wrote: Just a guess till the experts chime in: Linux disk I/O activity requires more CPU time than traditional Z Operating systems - so when one guest starts driving 5000 I/O ops per second to the swap device ( FBA mode vdisk in my case ) that in itself consumes a big chunk of CPU. Then there's the additional time spent in the linux kernel itself deciding what needs to go out to swap and what needs to come back in. let me re-emphasize this is a guess - I'd like to know the answer to this too.
Re: Need ideas for checking current terminal response time
TCPIP in z/VM has metrics that can be used to show network response time. z/VM provides transaction data. As a performance monitor that analyzes both, ESAMON will cost you less than 1% of a cpu. Could probably do what you want with just zMON at $1200/month. Horlick, Michael wrote: Greetings, Here is the situation. We are running z/VM 5.2 and 5 z/VSE 4.1.0 guest systems (3 production, 2 development machines) on an IBM 2066 (z800). In 2 of these VSEs there is a heavy duty CICS/TS system running. We use SET SHARE ABS to give them a minimum target of CPU, no limits, but sometimes I have to play around and give a hard limit to some VSEs when the system is slow and the CMS users (the programmers) call me complaining of response time. Sometimes it’s because within a production VSE virtual machine a batch job (or two or three) would be running. Anyways, I was thinking of somehow capturing what a CMS user response time would be every so often and perform some action (an alert or use the SET SHARE command) when the response is slow. I’m toying with creating a REXX EXEC which uses RXLDEV to create a logical 3270 session and have the EXEC basically “press” the ENTER key say every 30 seconds. I’m hoping this will mimic what a real interactive CMS user is experiencing. Take the time before and the time after with a ‘CP Q TIME’ and see how long it took. The question is how accurate would this be to the real thing (interactive CMS user doing “trivial” commands like XEDIT,etc…)? I do have CA-EXPLORE VM but I’m thinking that would be maybe more overhead in running and I am not sure that finding out the machine is running above, say 98% necessarily equates to a slow CMS response time. Would like your opinion, suggestions, etc… Thanks, Mike
Re: Need ideas for checking current terminal response time
ESAMON will let you know that minute. It analyzes the data every minute, alerts are set to check values every minute. So if that is after the fact, then yes Horlick, Michael wrote: Hi, Right now most users are still connected as SNA devices. We are slowly moving to TCP/IP. I assume that z/VM and ESAMON will let me know after the fact that there was slow interactive response during some interval(s) of time during the day. I would like something that would be more immediate (and of course, free ;)) Thanks, Mike -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Barton Robinson Sent: January 28, 2009 11:31 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Need ideas for checking current terminal response time TCPIP in z/VM has metrics that can be used to show network response time. z/VM provides transaction data. As a performance monitor that analyzes both, ESAMON will cost you less than 1% of a cpu. Could probably do what you want with just zMON at $1200/month. Horlick, Michael wrote: Greetings, Here is the situation. We are running z/VM 5.2 and 5 z/VSE 4.1.0 guest systems (3 production, 2 development machines) on an IBM 2066 (z800). In 2 of these VSEs there is a heavy duty CICS/TS system running. We use SET SHARE ABS to give them a minimum target of CPU, no limits, but sometimes I have to play around and give a hard limit to some VSEs when the system is slow and the CMS users (the programmers) call me complaining of response time. Sometimes it's because within a production VSE virtual machine a batch job (or two or three) would be running. Anyways, I was thinking of somehow capturing what a CMS user response time would be every so often and perform some action (an alert or use the SET SHARE command) when the response is slow. I'm toying with creating a REXX EXEC which uses RXLDEV to create a logical 3270 session and have the EXEC basically press the ENTER key say every 30 seconds. I'm hoping this will mimic what a real interactive CMS user is experiencing. Take the time before and the time after with a 'CP Q TIME' and see how long it took. The question is how accurate would this be to the real thing (interactive CMS user doing trivial commands like XEDIT,etc...)? I do have CA-EXPLORE VM but I'm thinking that would be maybe more overhead in running and I am not sure that finding out the machine is running above, say 98% necessarily equates to a slow CMS response time. Would like your opinion, suggestions, etc... Thanks, Mike
zPRO (tm) Product Announcement
Velocity Software is announcing zPRO, a portal for z/VM systems management. Functionality includes provisioning/cloning, as well as interfaces to many systems management functions for z/VM. More details can be found at http://velocitysoftware.com/zpro.html;. zPRO will be put up soon on our demonstration system and allow visitors to create their own servers - and hopefully to even logon to them to validate the process worked. zPRO is a long time objective of Velocity Software. As z/VM competes with other fully web enabled virtualization platforms, zPRO's objective is to provide z/VM with equivalent functionality and interface, fully leveraging z/VM functionality.
Re: VDISK
SYSTEM as displayed by ESALPS components is not system as in SYSTEM VMDBLK. System is really system totals. Which screen exactly are you looking at? (There isn't a vdisk storage by user display). The ESAASPC shows all the address spaces including VDISK, and shows 'SYSTEM'. The ESAVDSK just shows the active vdisks, sorted alphabetically by user, not by largest user, and should NOT show System. Schuh, Richard wrote: In looking at the VDISK Storage by User display (ESALPS) I noticed that SYSTEM was the 10th largest user of VDISK during one interval, but fell to 0 the next. What causes SYSTEM to show up as a user of VDISK? Perhaps something in the DEFINE or DETACH processes? Regards, Richard Schuh
Re: Web servers for VM
ESAWEB is continously being enhanced - and if you are looking at Linux is included as part of our ESALPS (Linux Performance Suite). The LINUXVM.ORG website runs on ESAWEB, as does VelocitySoftware.COM, and other websites as well. ESALPS was written in assembler to be fast. With lack of development on the other VM based webservers, we also have been providing migration tools from the other products to ESAWEB. ESAWEB is a large part of our long term strategy. If you already have other Velocity products, ESAWEB integrates in many ways to put web front end to your performance and other adminstrative tasks. Gentry, Stephen wrote: I'm wanting to know if there are any other web servers, commercial or other wise, available for VM. I know of the following products: VM:Webgateway (CA - no new development) ESAWEB (Velocity) z-Web-Server (Illustro) freeware/open source: httpd server from vm download page Thanks, Steve
Re: Web servers for VM
All conversion tools and some amount of conversion assistance is part of ESAWEB. Schuh, Richard wrote: Barton, Are the migration tools included with ESAWEB or are they available as separate packages? Regards, Richard Schuh -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Barton Robinson Sent: Tuesday, December 02, 2008 8:45 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Web servers for VM ESAWEB is continously being enhanced - and if you are looking at Linux is included as part of our ESALPS (Linux Performance Suite). The LINUXVM.ORG website runs on ESAWEB, as does VelocitySoftware.COM, and other websites as well. ESALPS was written in assembler to be fast. With lack of development on the other VM based webservers, we also have been providing migration tools from the other products to ESAWEB. ESAWEB is a large part of our long term strategy. If you already have other Velocity products, ESAWEB integrates in many ways to put web front end to your performance and other adminstrative tasks. Gentry, Stephen wrote: I'm wanting to know if there are any other web servers, commercial or other wise, available for VM. I know of the following products: VM:Webgateway (CA - no new development) ESAWEB (Velocity) z-Web-Server (Illustro) freeware/open source: httpd server from vm download page Thanks, Steve
Re: How can we control how much CPU is used by each zLinux guest?
Look at CP SET SHARE userid REL 100 ABS 5% LIMITSOFT will allow a linux server to only use 5% of the system unless no other user is ready to use CPU. Juarez, David T. wrote: What controls can be put in place to manage zLinux guests running under z/VM 5.3, so they do not saturate CPU and memory? Does the USER MSTOR parm really limit the zLinux guest to the amount coded? Thanks. David T. Juárez IT Specialist
Re: Page Space
Do the math Number one reason for ONE outage at each new z/linux installation is to fill up page space - guess you were lucky and had some extra spool space (no block paging so slow), so you luckily didn't take the outage - which makes your servers even slower Schuh, Richard wrote: Don't presume. 92G real, 10 xstore. All MDC activity is in real, limited to 384MB. And I do not know the color of the machine :-) Regards, Richard Schuh -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Marcy Cortes Sent: Thursday, November 13, 2008 10:29 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Page Space You didn't say how much real memory you have. Presumably less than 60G :) You either add enough real memory or you add enough page space to hold them all (at less that 50% occupied. I don't think there are miracles available in this scenario. Marcy This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose, or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation. From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Schuh, Richard Sent: Thursday, November 13, 2008 10:20 AM To: IBMVM@LISTSERV.UARK.EDU Subject: [IBMVM] Page Space Yesterday, we were running a test using 17 z/TPF virtual machines, 3GB each. This was in addition to the normal load on the system. During the test, which was not moving along very quickly, nothing was, I noticed that our page packs were 100% allocated, up from the usual 10%. This stood out as a smoking gun, verified by watching the performance improve as each of the ids in the test logged off. I presume that this should have been expected; however, other matters have kept us so busy that we did not do the math. I imagine that the one way to avoid this type of problem, we expect a peak of approximately 150 concurrent z/TPF systems in the coming year, is a massive injection of paging DASD. Is this the only answer or are there any other steps that we can take to help? Regards, Richard Schuh
Re: Page Space
Nah, mixing device types won't hurt much. as they fill, they start performing worse, and vm balances the load to ensure optiomal performance. read about mload. Mark Pace wrote: I have to find the stuff, but at the z Expo we were told that mixing DASD types in a page farm is BD! I don't remember right off hand why. Maybe someone else can chime in while I research. On Thu, Nov 13, 2008 at 3:28 PM, Schuh, Richard [EMAIL PROTECTED] wrote: What will be the effect, other than having additional space available, of adding five mod 9 disks to the existing page farm of 35 mod 3s? Would there be a noticeable change in the performance of the paging subsystem? (I suspect that any change will be less noticeable than the effects of filling both page and spool. :-) ) Regards, Richard Schuh -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Barton Robinson Sent: Thursday, November 13, 2008 11:36 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Page Space Do the math Number one reason for ONE outage at each new z/linux installation is to fill up page space - guess you were lucky and had some extra spool space (no block paging so slow), so you luckily didn't take the outage - which makes your servers even slower Schuh, Richard wrote: Don't presume. 92G real, 10 xstore. All MDC activity is in real, limited to 384MB. And I do not know the color of the machine :-) Regards, Richard Schuh -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Marcy Cortes Sent: Thursday, November 13, 2008 10:29 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Page Space You didn't say how much real memory you have. Presumably less than 60G :) You either add enough real memory or you add enough page space to hold them all (at less that 50% occupied. I don't think there are miracles available in this scenario. Marcy This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose, or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation. From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Schuh, Richard Sent: Thursday, November 13, 2008 10:20 AM To: IBMVM@LISTSERV.UARK.EDU Subject: [IBMVM] Page Space Yesterday, we were running a test using 17 z/TPF virtual machines, 3GB each. This was in addition to the normal load on the system. During the test, which was not moving along very quickly, nothing was, I noticed that our page packs were 100% allocated, up from the usual 10%. This stood out as a smoking gun, verified by watching the performance improve as each of the ids in the test logged off. I presume that this should have been expected; however, other matters have kept us so busy that we did not do the math. I imagine that the one way to avoid this type of problem, we expect a peak of approximately 150 concurrent z/TPF systems in the coming year, is a massive injection of paging DASD. Is this the only answer or are there any other steps that we can take to help? Regards, Richard Schuh
Re: Value added by z/VM versus VMWARE
Exactly, but the issue is to explain this to peter principal IT managers. Paul Raulerson wrote: I am very confused indeed by this whole conversation -VMWARE and z/VM solve different solutions. And they are both extraordinarily good at what they do. Just at the 10,000 foot level, VMWARE is designed to virtualize PC hardware and z/VM virtualizes mainframe hardware. Dismissing this as just two different hardware platforms is rather disingenuous, though admittedly, it is a true statement. Then again, a nuclear powered aircraft carrier and diesel powered megaton oil tanker are both ships - just two different hardware platforms. They hardly operate in the same realms though. Where everything starts to get different is the underlying hardware. And at that level, it gets very VERY different indeed. In some ways, VMWARE is more like an LPAR than a VM guest instance, but that difference is driven more by the hardware capabilities than by the design. -Paul On Nov 1, 2008, at 6:34 PM, Barton Robinson wrote: One thing that really bothers me about VMWARE. When I ask about performance to the people that measure, they tell me the VMWARE contract specifically states they are not allowed to talk about it's performance. A vendor that won't let people talk about performance must be very afraid details will be made public and don't really need to invest in improving it's performance. Since we can not provide facts to confuse management, it comes down to religion or companies providing their own facts. A professor from I think Stutgaart presented last year at the GSE/ IBM meeting pretty convincingly that VMWARE was about 20 years behind z/VM in almost any fair technological aspect you wish to evaluate. And I think he was wrong - I don't see sharing of resources in VMWARE even what z/VM had 20 years ago. VMWARE is much more like LPAR, so any argument you can use for z/VM vs LPAR works as well. I believe VMWARE is great for desktops where users may want to run applications that only run on different versions of windows or Linux. Now there is a company in California that is even virtualizing the desktops, give end users a small appliance, keyboard and monitor, and the software runs on a virtualized PC, where all software runs on the central virtualized PC that then supports multiple users. They save a lot of money by only having one copy of MS Office to support multiple end users. (Does this sound like 3270 and mainframes to anyone else?) Alan Ackerman wrote: Another question from the same architecture person. What is the value add ed by z/VM over VMWARE for a Linux workload? (That's my wording, not his.) As usual, I don't know anything about what VMWARE can or cannot do. I'm s ure it can run fewer guests than VM, but not how many. VM has shared DASD and DCSSes and NSSes , but most Linux people don't see the value of those things -- disks are cheap and come wi th the PC, memory is cheap, etc. VM has automation capabilities, but Linux has those too, and IBM sells all those Tivoli products to tie them together, report performance, provide high availabil ity, etc. I think the advantage on the mainframe is economy of scale. But how do yo u measure that? At present, you can save money on software and peripherals enough to cost -justify the mainframe. Reduced people costs are hard to quantify and scare the heck o ut of the midrange folks. But I wonder how long those software prices will last? Red Hat charges $1 8,000 per IFL for 7x24 support. (I found that on a web site, and I asked our Red Hat representat ive to make sure.) I couldn't find any prices on Novel SuSEs web site. We have other software with higher prices per engine for the mainframe. He specifically mentioned the ability to pick up a Linux guest running un der VMWARE and moving it to another box running VMWARE. So far VM cannot do that. Ideas on what value z/VM adds would be appreciated! Alan Ackerman Alan (dot) Ackerman (at) Bank of America (dot) com
Re: Value added by z/VM versus VMWARE
One thing that really bothers me about VMWARE. When I ask about performance to the people that measure, they tell me the VMWARE contract specifically states they are not allowed to talk about it's performance. A vendor that won't let people talk about performance must be very afraid details will be made public and don't really need to invest in improving it's performance. Since we can not provide facts to confuse management, it comes down to religion or companies providing their own facts. A professor from I think Stutgaart presented last year at the GSE/IBM meeting pretty convincingly that VMWARE was about 20 years behind z/VM in almost any fair technological aspect you wish to evaluate. And I think he was wrong - I don't see sharing of resources in VMWARE even what z/VM had 20 years ago. VMWARE is much more like LPAR, so any argument you can use for z/VM vs LPAR works as well. I believe VMWARE is great for desktops where users may want to run applications that only run on different versions of windows or Linux. Now there is a company in California that is even virtualizing the desktops, give end users a small appliance, keyboard and monitor, and the software runs on a virtualized PC, where all software runs on the central virtualized PC that then supports multiple users. They save a lot of money by only having one copy of MS Office to support multiple end users. (Does this sound like 3270 and mainframes to anyone else?) Alan Ackerman wrote: Another question from the same architecture person. What is the value add ed by z/VM over VMWARE for a Linux workload? (That's my wording, not his.) As usual, I don't know anything about what VMWARE can or cannot do. I'm s ure it can run fewer guests than VM, but not how many. VM has shared DASD and DCSSes and NSSes , but most Linux people don't see the value of those things -- disks are cheap and come wi th the PC, memory is cheap, etc. VM has automation capabilities, but Linux has those too, and IBM sells all those Tivoli products to tie them together, report performance, provide high availabil ity, etc. I think the advantage on the mainframe is economy of scale. But how do yo u measure that? At present, you can save money on software and peripherals enough to cost -justify the mainframe. Reduced people costs are hard to quantify and scare the heck o ut of the midrange folks. But I wonder how long those software prices will last? Red Hat charges $1 8,000 per IFL for 7x24 support. (I found that on a web site, and I asked our Red Hat representat ive to make sure.) I couldn't find any prices on Novel SuSEs web site. We have other software with higher prices per engine for the mainframe. He specifically mentioned the ability to pick up a Linux guest running un der VMWARE and moving it to another box running VMWARE. So far VM cannot do that. Ideas on what value z/VM adds would be appreciated! Alan Ackerman Alan (dot) Ackerman (at) Bank of America (dot) com
Re: Cost accounting for Linux guest running under z/VM
The best model is one I heard last week at the ibm conference. One large well known installation charges only for prime shift, with 3 different rates based on types of service. All charges are based on resource consumption. With the usage charges and prime shift only charges, users are convinced to move some batch operations off shift where resources are plentiful and free. This encourages high CPU activites that can be off shift to be moved off shift. The three different service rates relate to 7x24 support, prime shift support, and a best effort. Installations not charging for resource consumption in a mainframe environment tell their customers that tuning is not important, and neither is workload planning. This is good for IBM's profit as it increases the IFL requirement, but will end up with applications consuming more resource and can make other platforms much more attractive. One performance person is worth their power consumption in IFLs so to speak Most important is to understand your objective: chargeback and recover costs, or manage costs with chargeback and minimize resource requirements. Several installations have charged by server, and quickly found users will abuse the system unless encouraged otherwise. Juarez, David T. wrote: We are in the process of getting ready to add production Linux guest and we need to know how you are charging back the customer for running Linux under z/VM. We currently charge a fixed fee per month for small, medium and large based on the size of the individual server's memory, cpu, and network connections. Any ideas would be greatly appreciated. Thanks. David Juárez Department of Veterans Affairs IT Specialist - z/OS and z/VM Systems Programmer 512-326-6116
Re: Performance question
Best practices is to use Vdisk for swap, and reduce linux virtual machine sizes - not to buy more REAL z/VM memory unless you really need it. 101% memory useage means almost nothing. It is not relevant to performance or capacity, and thus shouldn't have business decisions or performance decisions decided based on that number. Dean, David (I/S) wrote: My SLES 10.1 zLinux servers have been notorious for not making much use of the swap space, even when we lower the main (virtualized) in the USER DIRECTORY. We may have something just not configured optimally somewhere, but our luck has gone with adding main Linux memory and subsequently having to add ZVM memory. David Dean Information Systems *bcbstauthorized* From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Mary Anne Matyaz Sent: Monday, September 29, 2008 11:31 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Performance question If your VM system is at 101% memory usage, and you are overcommitted by about 14%, is it worthwhile to add a vdisk to a linux for swap space, or better just to add main memory to the linux? MA (Looking for opinions, thoughts, rationalizations, whatever. :) Please see the following link for the BlueCross BlueShield of Tennessee E-mail disclaimer: http://www.bcbst.com/email_disclaimer.shtm
Re: Performance question
z/VM Memory usage, what do you think it means? If a page of a virtual machine is in storage, but has not been referenced in 10 minutes, is that part of your percent used? Likely you don't know the answer and the source of your information doesn't either. So if that's the case, what information are you using to make decisions? Mary Anne Matyaz wrote: I know I'm probably going to regret this, but, how can that be? I said VM memory usage, right? Not Linux MA On Mon, Sep 29, 2008 at 1:23 PM, Barton Robinson [EMAIL PROTECTED] wrote: Best practices is to use Vdisk for swap, and reduce linux virtual machine sizes - not to buy more REAL z/VM memory unless you really need it. 101% memory useage means almost nothing. It is not relevant to performance or capacity, and thus shouldn't have business decisions or performance decisions decided based on that number.
Re: Performance question
Best practices is to use Vdisk for swap Mary Anne Matyaz wrote: If the page has not been referenced in 10 minutes, but is not paged out, I would expect it to be included in the 101%. Try not to focus so much on the extraneous info and address the question, if I am using a huge amount of memory, is it more helpful to use vdisk or guest memory? MA On Mon, Sep 29, 2008 at 3:32 PM, Barton Robinson [EMAIL PROTECTED] wrote: z/VM Memory usage, what do you think it means? If a page of a virtual machine is in storage, but has not been referenced in 10 minutes, is that part of your percent used? Likely you don't know the answer and the source of your information doesn't either. So if that's the case, what information are you using to make decisions? Mary Anne Matyaz wrote: I know I'm probably going to regret this, but, how can that be? I said VM memory usage, right? Not Linux MA On Mon, Sep 29, 2008 at 1:23 PM, Barton Robinson [EMAIL PROTECTED] wrote: Best practices is to use Vdisk for swap, and reduce linux virtual machine sizes - not to buy more REAL z/VM memory unless you really need it. 101% memory useage means almost nothing. It is not relevant to performance or capacity, and thus shouldn't have business decisions or performance decisions decided based on that number.
VM:Webgateway to ESAWEB Conversion Tool
We've been asked to provide such a tool by some of our customers. Please see details about ESAWEB (http://velocitysoftware.com/esaweb.html;) and a link to details on the conversion tool if you would be interested in such a tool.
Re: IUCV - What's wrong with this picture?
Sounds like there is a need for decent performance monitoring. dave wrote: Hi, Gary. Well, there is no such thing as a free lunch, so establishing *large* numbers of IUCV connections between virtual machines does cost something. Control blocks must be allocated, must be managed by CP, interrupts fielded, etc. Off of the top of my head, I don't know how much storage these control blocks take, but I would suspect that with CP now being 64-bit, the amount of storage taken would not be a significant issue. Even if the amount of traffic between the clients and the VM server is slight; the *timing* of the traffic might be a concern.5000 clients all sending a short IUCV message at the same time to the server, might cause problems. The server would have to have enough resources available to process all of the traffic in an acceptable amount of time Good luck. - Original Message - From: Gary M. Dennis [EMAIL PROTECTED] To: IBMVM@LISTSERV.UARK.EDU Subject: IUCV - What's wrong with this picture? Date: Mon, 25 Aug 2008 12:23:49 -0500 Assumptions: 0. A VM server machine 1. A cluster of client virtual machines (possibly thousands) 2. n buffers are allocated for each client virtual machine 3. Each buffer contains table elements that require (a) Element ageing (b) Element deletion when invalidated by: 1. lack of use 2. client machine request (c) Compression as buffer fragmentation occurs 4. Each client virtual machine in the cluster is connected via IUCV to the server virtual machine. 5. IUCV traffic between the server machine and client machine is extremely low volume. Initial call, termination call, intermittent statistics call. 6. After the initial call, the server virtual machine will maintain the buffer table entries in each client virtual machine without additional IUCV interaction. Now the questions: 1. Does IUCV infrastructure overhead specifically associated with number of connections become prohibitive at some well known point? 2. Has anyone had experience with an application having a high IUCV connection count like this? If so, what was that experience? Again, the traffic incidence per connection is very low but the number of connections is potentially very high. Thanks --. .- .-. -.-- Gary Dennis
Re: z/VM 5.2 and the 2GB Line
140GB didn't leave room for work, and make sure you have lots more expanded. Schuh, Richard wrote: Let's say that total amount of virtual storage is the main issue, with everything else relegated to the status of being inconsequential. Regards, Richard Schuh -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Bill Holder Sent: Thursday, August 21, 2008 1:54 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: z/VM 5.2 and the 2GB Line On Thu, 21 Aug 2008 09:57:22 -0700, Schuh, Richard [EMAIL PROTECTED] wrot= e: I have been unable to come up with the magic search argument to get this= list to spit out the answer to the question, What is the total GB of virtual storage, the sum of the VM sizes, can z/VM 5.2 handle without running into the dreaded 2GB line constraint?, so I will ask the list. What is it? Regards, Richard Schuh The answer is, of course, the trademarked it depends. There are severa= l different forms of the constraint, and all are dependent on workload. Ev= en something as apparently unrelated as how dense or sparse the guests' stor= age reference patterns tend to be can greatly influence the degree to which t= he constraint will be an issue for a given total virtual storage size. I'm = not sure if we even have typical or average numbers, but I can ask around. = - Bill Holder, z/VM Development, IBM
Re: DOS attack details in
The port and IP address sending the request should be in the monitor records. There would some inforamation useful there. Mike Walter wrote: Back on July 15, we experienced our first known Denial of Service attack (more likely a problem server). I reported it to our Internet Security group including: From the nearly anonymous/invisible TCPIPMESSAGE file in TCPMAINT's reader: ---snip DTCUTI001E Serious problem encountered: 15:38:55 07/15/08 DTCUTI002E A denial-of-service attack has been detected ---snip--- Issued after the nearly anonymous/invisible TCPIPMESSAGE file in TCPMAINT's reader was accidentally discovered: ---snip--- netstat dos VM TCP/IP Netstat Level 510 Maximum Number of Half Open Connections: 512 Denial of service attacks: Attacks Elapsed Attack Attack IP Address Detected Time Duration --- - - - Smurf-IC 10.64.103.250 1 2:27:08 0:00:00 Ready; T=0.02/0.02 18:13:13 ---snip--- So I asked our Internet Security team who might be the offending 10.64.103.250. In turn they asked me for the port number being used for this attack, and the mac address of the attacking machine. Unfortunately, none of that is available after the attack (which was admirably and automatically quashed by the z/VM TCPIP stack). Would it be possible to include more information in the nearly anonymous/invisible TCPIPMESSAGE file in TCPMAINT's reader, including the port being used and the MAC address, and the other information displayed by the NETSTAT DOS command? If the attack is discovered after the next time the stack is restarted, NETSTAT DOS doesn't provide any information. Actually, I don't see any reason why all that information could not be logged to the TCPIP stack console itself - as a single point of reference should an investigation be required later. BTW, the current release of VM:Operator loops (or otherwise fails to ever respond) when the NETSTAT command is issued, so we can't even issue an automated NETSTAT DOS command, trap the response, and try to gather useful information during the attack. Mike Walter Hewitt Associates Any opinions expressed herein are mine alone and do not necessarily represent the opinions or policies of Hewitt Associates. The information contained in this e-mail and any accompanying documents may contain information that is confidential or otherwise protected from disclosure. If you are not the intended recipient of this message, or if this message has been addressed to you in error, please immediately alert the sender by reply e-mail and then delete this message, including any attachments. Any dissemination, distribution or other use of the contents of this message by anyone other than the intended recipient is strictly prohibited. All messages sent to and from this e-mail address may be monitored as permitted by applicable law and regulations to ensure compliance with our internal policies and to protect our business. E-mails are not secure and cannot be guaranteed to be error free as they can be intercepted, amended, lost or destroyed, or contain viruses. You are deemed to have accepted these risks if you communicate with us by e-mail.
SHARE with us Velocity Software's 20th Anniversary
The SHARE meeting was graciously located in San Jose, California for Velocity Software's 20th Anniversary. Velocity Software was incorporated 8/8/88 close by in Mountain View, CA. We have been a very active part of VM/XA, VM/ESA and now z/VM in the UPs, DOWNs and now back UPs of the last 20 years. As part of our anniversary present to ourselves, we celebrate our new z9 where you will see the LinuxVM.ORG website start to run much faster, and our Linux performance management demonstrations fly. If you happen to speak German, you might notice we've expanded with Velocity Software GmbH with the website found at VelocitySoftware.DE. I would like to invite all friends of Velocity Software to an event thursday night, August 14th. This event will include some of my favorite beverages and food at one of my favorite wineries in the area. Space is limited (by the winery), please see Cheri at our booth at the EXPO area at SHARE for your ticket. We will provide transportation.
Re: Nice idea in blog: Should we toss x86 architecture - NOT.
If so, then unlikely that CMS would run on cell blade engines, and emulation not required. With IBM now owning platform, who did seem to have this kind of technology, there are feasible options that would actually be marketable. Quay, Jonathan (IHG) wrote: It is my understanding that IBM intends to integrate Cell Blade engine (e.g. playstation 3) technology into the z/Series ecosystem. This would seem to me to be the place where massively parallel high intensity cpu workload would live in the not so far flung future. -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Barton Robinson Sent: Wednesday, July 23, 2008 10:59 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Nice idea in blog: Should we toss x86 architecture - NOT. Ok, so reality check folks before y'all start drooling about jobs and can think you can run 47000 windows servers under VM. In Linux we learned that running compiled code natively on z, megahertz is megahertz and a CPU intensive task would always run faster on Intel than on z (until we got z9 and z10). And that is native meaning the programs were compiled to run on z, and the operating system was compiled to run on z. So now, under CMS, this emulates intel. So megahertz is NOT megahertz. With emulating an architecture, one could easily imagine losing an order of magnitude. Thus a windows server that is running at 10% peak on a 4Ghz processor would consume a z10 IFL and want more. One does need to pay significant attention to the performance characteristics before thinking about something like this seriously. Sorry. Gary M. Dennis wrote: Z/VOS is a CMS application. The glass-side user will only see Windows via RDC and know nothing of or about CMS or VM. Gary On 7/22/08 8:30 PM, dave [EMAIL PROTECTED] wrote: Good luck, Gary. I do hope your organization can pull this off. VM-ers need more employment possibilities:-) I gather from some of your previous posts to this list that your Windows support software, z/VOS, is in fact a sophisticated CMS-based application, that is a user would log onto a CMS user id to start his Windows systemis my understanding correct? Thanks and have a good one. DJ - Original Message - From: Gary M. Dennis [EMAIL PROTECTED] To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Nice idea in blog: Should we toss x86 architecture Date: Tue, 22 Jul 2008 13:02:33 -0500 This was our post to the zd net blog. Maybe we already have. In Q1 2009 Mantissa will deliver a system that permits unaltered Windows operating systems to run under z/VM. Using a desktop appliance running RDC, users will be able to connect to their virtual Windows images running in the VM environment. Goodbye desktop hardware, remote maintenance, high power consumption, machine order lead time. z/VOS began with the observation that most Windows workstations do practically nothing 95% of the time and we were so intrigued with the idea of being able to actually run an intel-based operating system under IBM VM that we never looked back. VM provided a natural platform for development of this product. The product has been a bear for the development group but the thought of being able to run 3000 copies of Windows on one System z so fascinated the team that we needed very little additional incentive. Let's hope IBM can ramp up System z production. Why wait until 2016? --. .- .-. -.-- Gary Dennis Mantissa Corporation On 7/22/08 11:14 AM, Bob Heerdink [EMAIL PROTECTED] wrote: http://blogs.zdnet.com/perlow/?p=9183 Should we toss x86 architecture and wipe the slate with something greene r and more scalable? Windows Server 2016 128-bit edition running virtualized on z/VM in a gre en datacenter, accessed via my house from a thin client over high-speed fibe r optic connection. I can see it now. Hope this happens sooner than predicted, Bob
Re: Nice idea in blog: Should we toss x86 architecture - NOT.
Ok, so reality check folks before y'all start drooling about jobs and can think you can run 47000 windows servers under VM. In Linux we learned that running compiled code natively on z, megahertz is megahertz and a CPU intensive task would always run faster on Intel than on z (until we got z9 and z10). And that is native meaning the programs were compiled to run on z, and the operating system was compiled to run on z. So now, under CMS, this emulates intel. So megahertz is NOT megahertz. With emulating an architecture, one could easily imagine losing an order of magnitude. Thus a windows server that is running at 10% peak on a 4Ghz processor would consume a z10 IFL and want more. One does need to pay significant attention to the performance characteristics before thinking about something like this seriously. Sorry. Gary M. Dennis wrote: Z/VOS is a CMS application. The glass-side user will only see Windows via RDC and know nothing of or about CMS or VM. Gary On 7/22/08 8:30 PM, dave [EMAIL PROTECTED] wrote: Good luck, Gary. I do hope your organization can pull this off. VM-ers need more employment possibilities:-) I gather from some of your previous posts to this list that your Windows support software, z/VOS, is in fact a sophisticated CMS-based application, that is a user would log onto a CMS user id to start his Windows systemis my understanding correct? Thanks and have a good one. DJ - Original Message - From: Gary M. Dennis [EMAIL PROTECTED] To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Nice idea in blog: Should we toss x86 architecture Date: Tue, 22 Jul 2008 13:02:33 -0500 This was our post to the zd net blog. Maybe we already have. In Q1 2009 Mantissa will deliver a system that permits unaltered Windows operating systems to run under z/VM. Using a desktop appliance running RDC, users will be able to connect to their virtual Windows images running in the VM environment. Goodbye desktop hardware, remote maintenance, high power consumption, machine order lead time. z/VOS began with the observation that most Windows workstations do practically nothing 95% of the time and we were so intrigued with the idea of being able to actually run an intel-based operating system under IBM VM that we never looked back. VM provided a natural platform for development of this product. The product has been a bear for the development group but the thought of being able to run 3000 copies of Windows on one System z so fascinated the team that we needed very little additional incentive. Let's hope IBM can ramp up System z production. Why wait until 2016? --. .- .-. -.-- Gary Dennis Mantissa Corporation On 7/22/08 11:14 AM, Bob Heerdink [EMAIL PROTECTED] wrote: http://blogs.zdnet.com/perlow/?p=9183 Should we toss x86 architecture and wipe the slate with something greene r and more scalable? Windows Server 2016 128-bit edition running virtualized on z/VM in a gre en datacenter, accessed via my house from a thin client over high-speed fibe r optic connection. I can see it now. Hope this happens sooner than predicted, Bob
Re: Bogus CPU utilization numbers from Linux Red Hat 4.6
Please. ESAMON has been correcting the Linux numbers since the problem was discovered in 2001. Thomas Kern wrote: I think that the discussion was that tools like PERFTK, ESAMON, CP IND USER show accurate numbers for what the whole virtual machine is using, and the the numbers from tools INSIDE a linux virtual machine such as TOP, SAR vary depending upon the level of the kernel, the distribution and the workload of the rest of the system. /Tom Kern Martin, Terry R. (CMS/CTR) (CTR) wrote: Hi I saw something in one of the postings that stated that the CPU utilization numbers that were reported in the z/VM Performance Toolkit on behalf of a Red Hat 4.6 z/Linux guest were not correct. Is it only the PTK that does not report the correct numbers or is it any monitor? Do we know if the numbers reported are bogus on the high side or low side? I am running z/VM 5.3 and the Linux Kernel is at 2.6.9-67. Thanks Terry
Re: Second Physical Screen for Performance Monitor
Not really the monitor, but the performance monitor ESAMON does that standard. Howard Rifkind wrote: Off the top of the lists hat would anyone know if you can connect a second physical monitor for z/VM. I would like to have one in the computer room and one in the Systems Programmers area. I know you can have a web interface to the monitor but that isn't way the manager wants. Thanks. _ LEGAL NOTICE Unless expressly stated otherwise, this message is confidential and may be privileged. It is intended for the addressee(s) only. Access to this E-mail by anyone else is unauthorized. If you are not an addressee, any disclosure or copying of the contents of this E-mail or any action taken (or not taken) in reliance on it is unauthorized and may be unlawful. If you are not an addressee, please inform the sender immediately, then delete this message and empty from your trash.
Re: Monitor for zVM
zMON at $1200/year? Huegel, Thomas wrote: Depending on what you need HOBIT (a freebe) may work for you. -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] Behalf Of LOREN CHARNLEY Sent: Tuesday, June 03, 2008 8:59 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Monitor for zVM I have been mandated by management to seek a lower cost performance monitor than our currently employed ESAMON from Velocity Software. The reason for this is that we are in the throws of down sizing what we are running on our current z800 with zVM 4.3. There is also talk of eliminating zVM altogether, but that is a discussion for a later date. If anyone can suggest something of zero cost or at least lower than ESAMON, I would appreciate it. TIA, Loren Charnley, Jr. IT Systems Engineer FAMILY DOLLAR (704) 847-6961 Ext. 3327 (704) 814-3327 [EMAIL PROTECTED]
Re: BEST /1 from BMC
I'm not Phil, but no. Our data layouts are published, so VISUALIZER could if BMC wanted. Bill Munson wrote: Phil, Is anyone that is using the ESAMAP History files as input to there VISUALIZER getting the same numbers out ? thanx Bill Munson VM System Programmer 201-418-7588 Phil Smith III [EMAIL PROTECTED] Sent by: The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU 05/16/2008 06:09 AM Please respond to The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU To IBMVM@LISTSERV.UARK.EDU cc Subject Re: BEST /1 from BMC Is there anyone using UIE/VM the BEST/1 product from BMC ? Obviously I'm not, but at least one other Velocity customer was. What's the question? ...phsiii
Re: Overcommit ratio
Stephen, you are doing great. Your workload must be Oracle, and not WAS, DB2 or Domino. If it is WAS, it must be old prior to performance enhancements. so don't upgrade it. And the metric IS useful, you know if you add 4 more servers how much more mainframe storage you need. And your number gives a reference point to others to show what they could be doing if everything worked correctly. Stephen Frazier wrote: My overcommit ratio is about 5:1 not counting CMS users. If you count them it is more like 15:1. It seems to work fine. I don't think overcommit ratio is very useful for anything. It is two dependent on the kind of users you have to be meaningful. Marcy Cortes wrote: I keep hearing things like shouldn't be overcommitted in prod more than 2:1 or 3 or 4:1 in test. How is that calculated? Can I just take the (Pageable storage number + Pages on DASD ) / pageable storage number? Marcy This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose, or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation.
Re: Overcommit ratio
My use of the term over-commit is more simple with the objective of setting a target that management understands. I don't include vdisk - that is a moving target based on tuning and workload, as is the use of CMM1. The way I like to use the term is much higher level that doesn't change based on workload. I would use (Defined Guest Storage) / (CENTRAL + EXPANDED) (and people that use MDC indiscriminately or vise versa need some perforance assistance, but that is part of the tuning) With this, I have the objective of managing to this target. So using CMM (1) to reduce storage and the use of VDISK increases storage is the tuning part. And then I have a measurement that is compareable across systems - especially important when virtual technologies are competing and other virtual platforms don't/can't overcommit. This is a serious measure of technology and tuning ability as well. With current problems in JAVA/Websphere, Domino and some other Tivoli applications, I've seen the overcommit ratio attainable drop considerably. I used to expect 3 to 7 attainable, now some installations are barely able to attain 1.5. This starts to make VMWARE where 1 is a good target look better - not in our best interest. And it gives me a measure of an installation's skill set (or ability to tune based on tools of course). It would be interesting to get the numbers as i've defined for installations. Using this measure, what do y'all run? MARCY WROTE: Well, only if the server uses them. If you have a 1.5G server and it is using 1.5 Gig of swap space in VDISK then it is an impact of 3G virtual, right? If you have a 1.5G server and it is not swapping, it's impact is 1.5G virtual. So maybe more like (sum (guest virtual storage sizes) + sum (*used* vdisk blocks) ) / central storage. Wouldn't that be pretty simliar to number of pages on DASD method? Expanded storage? Add it to central? Nothing's simple anymore :) Marcy Cortes Rob van der Heij wrote: On Tue, May 13, 2008 at 6:02 AM, Robert J Brenneman [EMAIL PROTECTED] wrote: The problem will be when you've allocated huge vdisks for all your production systems based on the old Swap = 2X main memory ROT. In that example - you're basically tripling your overcommit ratio by including the vdisks. This also can have a large cost in terms of CP memory structures to manage those things. I think you are confusing some things. In another universe there once was a restriction of *max* twice the main memory as swap, but that was with another operating system to start with. Linux needs swap space to allow over-commit within Linux itself. The amount of swap space is determined by the applications you run and their internal strategy to allocate virtual memory. That space is normally not used by Linux. The current guidance is a smallish vdisk for high priority swap space, and a largish low priority real disk/minidisk for occasional use by badly behaved apps. Swapping to the vdisk is fine in normal operations, swapping to the real disk should be unusual and rare. The unused swap disk should only be on real disk when you have no monitoring set up. In that case when Linux does use it, things get so slow that your users will call your manager to inform you about it. The VDISK for swap that is being used actively by Linux during peak periods is completely different. That's your tuning knob to differentiate between production and development servers, for example. It reduces the idle footprint of the server at the expense of a small overhead during the (less frequent) peak usage. That tuning determines the application latency and paging requirements. I believe the over-commit ratio is a very simplified view of z/VM memory management. It does not get much better by adding other factors. Just use the sum of virtual machine and VDISK. And remember to subtract any other things like MDC from your available main storage. Rob
Re: Overcommit ratio
Ah yes, CMS is very different animal - it knows how to work well in a virtual environment. I think I remember numbers way above 20, so high nobody bothered to measure. Huegel, Thomas wrote: My ratio is about 2.6 that represents a large (proportionately) number of CMS users. -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] Behalf Of Barton Robinson Sent: Tuesday, May 13, 2008 10:20 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Overcommit ratio My use of the term over-commit is more simple with the objective of setting a target that management understands. I don't include vdisk - that is a moving target based on tuning and workload, as is the use of CMM1. The way I like to use the term is much higher level that doesn't change based on workload. I would use (Defined Guest Storage) / (CENTRAL + EXPANDED) (and people that use MDC indiscriminately or vise versa need some perforance assistance, but that is part of the tuning) With this, I have the objective of managing to this target. So using CMM (1) to reduce storage and the use of VDISK increases storage is the tuning part. And then I have a measurement that is compareable across systems - especially important when virtual technologies are competing and other virtual platforms don't/can't overcommit. This is a serious measure of technology and tuning ability as well. With current problems in JAVA/Websphere, Domino and some other Tivoli applications, I've seen the overcommit ratio attainable drop considerably. I used to expect 3 to 7 attainable, now some installations are barely able to attain 1.5. This starts to make VMWARE where 1 is a good target look better - not in our best interest. And it gives me a measure of an installation's skill set (or ability to tune based on tools of course). It would be interesting to get the numbers as i've defined for installations. Using this measure, what do y'all run? MARCY WROTE: Well, only if the server uses them. If you have a 1.5G server and it is using 1.5 Gig of swap space in VDISK then it is an impact of 3G virtual, right? If you have a 1.5G server and it is not swapping, it's impact is 1.5G virtual. So maybe more like (sum (guest virtual storage sizes) + sum (*used* vdisk blocks) ) / central storage. Wouldn't that be pretty simliar to number of pages on DASD method? Expanded storage? Add it to central? Nothing's simple anymore :) Marcy Cortes Rob van der Heij wrote: On Tue, May 13, 2008 at 6:02 AM, Robert J Brenneman [EMAIL PROTECTED] wrote: The problem will be when you've allocated huge vdisks for all your production systems based on the old Swap = 2X main memory ROT. In that example - you're basically tripling your overcommit ratio by including the vdisks. This also can have a large cost in terms of CP memory structures to manage those things. I think you are confusing some things. In another universe there once was a restriction of *max* twice the main memory as swap, but that was with another operating system to start with. Linux needs swap space to allow over-commit within Linux itself. The amount of swap space is determined by the applications you run and their internal strategy to allocate virtual memory. That space is normally not used by Linux. The current guidance is a smallish vdisk for high priority swap space, and a largish low priority real disk/minidisk for occasional use by badly behaved apps. Swapping to the vdisk is fine in normal operations, swapping to the real disk should be unusual and rare. The unused swap disk should only be on real disk when you have no monitoring set up. In that case when Linux does use it, things get so slow that your users will call your manager to inform you about it. The VDISK for swap that is being used actively by Linux during peak periods is completely different. That's your tuning knob to differentiate between production and development servers, for example. It reduces the idle footprint of the server at the expense of a small overhead during the (less frequent) peak usage. That tuning determines the application latency and paging requirements. I believe the over-commit ratio is a very simplified view of z/VM memory management. It does not get much better by adding other factors. Just use the sum of virtual machine and VDISK. And remember to subtract any other things like MDC from your available main storage. Rob
Re: SNMP client for CMS
ESALPS provides this. Shedlock, George wrote: Does anyone know of a program or utility that can generate an SNMP message preferably from a REXX exec? George Shedlock Jr AEGON Information Technology AEGON USA 502-560-3541
Re: Using SET SHARE, performnace problem
So at times of peak CPU, you need to share the CPU to provide better CICS response times. Use SET SHARE vsebatch REL 100 ABS 30% LIMITSOFT. Large shares will NOT do what you think or want. This command lets the batch use default share, but caps it at 30% CPU unless there are no other users. If your CICS interactive and your batch are in the same server, then you need to prioritize within VSE. Horlick, Michael wrote: Cross-posted to both VMESA-L and VSE-L mailing lists Greetings, We have just converted the last of our 5 VSE machines to z/VSE 4.1.0 (from VSE/ESA 2.6.1) and are experiencing performance issues. My peak times are 98-100% utilization and people are complaining about poor response times. I don't know whether it's because I am using CICS data tables more now or because of the additional CPU utilization for z/VSE. Anyways, one question I have is the usage of the SET SHARE. I have been using the 'SHARE ABSOLUTE' directory control statement for each of my VSE machines (giving say 38% to one machine, giving 29% to another,etc...) with maximum share nolimit. The problems seem to occur when batch jobs are run in these predominately CICS/TS systems. I was wondering if maybe a SET SHARE RELATIVE technique would be more effective and what you do in prioritizing virtual machines within the physical machine? Thanks, Mike
Re: Using SET SHARE, performance problem
no no no no no CPU is the bottleneck, not queues, not paging. THis is a matter of CPU redistribution. or application tuning, or talk to your ibm business partner about more CPU. Mike Hammock wrote: Did you perhaps increase the size of the virtual machines when going to zVSE 4.1?? In any case, I'd check for an eligible list. (do #CP IND Q and look for any of your guests in E3). If any VSE guest is in E3, I'd suggest (carefully) adjusting the SRM STORBUFF setting to allow more overcommittment of real storage. Monitor your paging activity and page space usage carefully. Mike C. M. (Mike) Hammock Sr. Technical Support zFrame IBM zSeries Solutions (404) 643-3258 [EMAIL PROTECTED] Horlick, Michael michael.horlick@ To cgi.com IBMVM@LISTSERV.UARK.EDU Sent by: The IBM IBMVM@LISTSERV.UARK.EDU z/VM Operating cc System [EMAIL PROTECTED] Subject ARK.EDU Using SET SHARE, performnace problem 04/23/2008 01:38 PM Please respond to The IBM z/VM Operating System [EMAIL PROTECTED] ARK.EDU Cross-posted to both VMESA-L and VSE-L mailing lists Greetings, We have just converted the last of our 5 VSE machines to z/VSE 4.1.0 (from VSE/ESA 2.6.1) and are experiencing performance issues. My peak times are 98-100% utilization and people are complaining about poor response times. I don’t know whether it’s because I am using CICS data tables more now or because of the additional CPU utilization for z/VSE. Anyways, one question I have is the usage of the SET SHARE. I have been using the ‘SHARE ABSOLUTE’ directory control statement for each of my VSE machines (giving say 38% to one machine, giving 29% to another,etc…) with maximum share nolimit. The problems seem to occur when batch jobs are run in these predominately CICS/TS systems. I was wondering if maybe a SET SHARE RELATIVE technique would be more effective and what you do in prioritizing virtual machines within the physical machine? Thanks, Mike __ This email may contain confidential and privileged material for the sole use of the intended recipient(s). Any review, use, distribution or disclosure by others is strictly prohibited. If you are not the intended recipient (or authorized to receive for the recipient), please contact the sender by reply email and delete all copies of this message. To reply to our email administrator directly, send an email to [EMAIL PROTECTED]
Re: VTAM R.I.P.
Gee, next it will be the high cost of z/OS that you will be looking at. How much do you save if you move an application from z/OS to z/Linux? Colin Allinson wrote: Rob van der Heij [EMAIL PROTECTED] wrote :- Z NET,QUICK Couldn't be 'QUICK' enough for us. We managed to eliminate it from VM by the middle of last year. We went back to using basic mode CTC connections to z/OS until they got themselves up to 1.7 with the ability to use TCPNJE. The interesting thing was that it was the huge cost of VTAM that was the main motivation for us. Given that the product was functionally stabilised and needed virtually zero support for a number of years we were struggling to see why. Once we looked at VTAM, and eliminated it, this led us to look at a number of other relatively high cost products that we have ways to eliminate, replace or reduce. I am not saying that we would not have looked at these anyway but the high cost of VTAM was the catalyst to start looking. Colin Allinson Amadeus Data Processing GmbH
Re: VTAM R.I.P.
WOW, I really had no idea it would be this significant. No wonder IBM sales people don't sell Linux to replace z/OS. So conservatively, 90% reduction in costs for any application that moves? So about 5 mips (a p390 worth) would pay for any Velocity costs? Amazing. Said, Nick wrote: Our management came up with: Per MIP Cost Analysis Environment Onetime Ongoing z/OS $7,300 $1,980 z/Linux $447 $61 Of course, this does not include the cost of any Velocity Software products on z/VM :) -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Barton Robinson Sent: Thursday, April 03, 2008 12:38 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: VTAM R.I.P. Gee, next it will be the high cost of z/OS that you will be looking at. How much do you save if you move an application from z/OS to z/Linux? Colin Allinson wrote: Rob van der Heij [EMAIL PROTECTED] wrote :- Z NET,QUICK Couldn't be 'QUICK' enough for us. We managed to eliminate it from VM by the middle of last year. We went back to using basic mode CTC connections to z/OS until they got themselves up to 1.7 with the ability to use TCPNJE. The interesting thing was that it was the huge cost of VTAM that was the main motivation for us. Given that the product was functionally stabilised and needed virtually zero support for a number of years we were struggling to see why. Once we looked at VTAM, and eliminated it, this led us to look at a number of other relatively high cost products that we have ways to eliminate, replace or reduce. I am not saying that we would not have looked at these anyway but the high cost of VTAM was the catalyst to start looking. Colin Allinson Amadeus Data Processing GmbH This email is intended for the recipient only. If you are not the intended recipient please disregard, and do not use the information for any purpose.
Re: SHARE vs. zSeries Expo
And I paid for a vendor session to give that non-vendor presentation. Gee, SHARE encourages technical presentations, and EXPO has lots of sessions that are pretty poorly attended - as in little interest. Marcy Cortes wrote: Did both last year. While there is a lot of overlap, SHARE is a heavier on technical how'tos and user experiences sessions and non-IBM vendor content - Barton only got 1 session at expo :)... IMHO. You can use your IBM credits for all the z10's you're purchasing to attend Expo, though :) Marcy Cortes This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose, or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation. -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of O'Brien, Dennis L Sent: Wednesday, March 26, 2008 1:35 PM To: IBMVM@LISTSERV.UARK.EDU Subject: [IBMVM] SHARE vs. zSeries Expo We're planning our training for the year, and wondering about the value of SHARE vs. zSeries Expo. Several of us have been to SHARE, but none have been to the Expo. What do people who've been to both think of each? Dennis O'Brien Just because we spent the night together doesn't mean we're on a first name basis. -- Miss Glick, in Lucky Stiff
Re: Performance problem Linux under Zvm
The guideline for LDUBUF is to LOWER it from default, NEVER raise it unless you like to re-IPL z/VM. O'Brien, David W. (NIH/CIT) [C] wrote: I made the mistake of believing what I was told. cp q storage 05:57:21 STORAGE = 2G Ready; T=0.01/0.01 05:57:21 cp q virtual storage 05:57:35 STORAGE = 128M Ready; T=0.01/0.01 05:57:35 cp q srm 06:01:42 IABIAS : INTENSITY=90%; DURATION=2 06:01:42 LDUBUF : Q1=100% Q2=75% Q3=60% 06:01:42 STORBUF: Q1=300% Q2=300% Q3=300% 06:01:42 DSPBUF : Q1=32767 Q2=32767 Q3=32767 06:01:42 DISPATCHING MINOR TIMESLICE = 5 MS 06:01:42 MAXWSS : LIMIT=% 06:01:42 .. : PAGES=99 06:01:42 XSTORE : 0% Any recommendation for LDUBUF? I just raised it to 100 100 100. Please advise if that change was counter indicated. Thank you, Dave O'Brien National Institutes of Health From: Gentry, Stephen [Sent: Thu 3/20/2008 2:51 PM Subject: Re: Performance problem Linux under Zvm What command did you use to determine that you had 768m central storage? QUERY STOREAGE? QUERY VIRTUAL STORAGE? Steve G. -Original Message- Subject: Re: Performance problem Linux under Zvm Thanks John cp q srm IABIAS : INTENSITY=90%; DURATION=2 LDUBUF : Q1=100% Q2=75% Q3=60% STORBUF: Q1=300% Q2=200% Q3=200% DSPBUF : Q1=32767 Q2=32767 Q3=32767 DISPATCHING MINOR TIMESLICE = 5 MS MAXWSS : LIMIT=% .. : PAGES=99 XSTORE : 0% Just got the following from one of the other techs (non-VM) We were able to diagnose the problem and make the necessary correction. The problem was z/VM has a total 768m of central available. The Linux guests (3 total) each had 768m of central allocated, therefore contention. The Linux guests are over allocated and are storage constrained with 768m of central. Understanding the Linux guests would be in contention with each other for this storage VM time sliced what it could for each guest, therefore the symptoms we experienced. My question to this group - Does a Linux quest really require 768MB of Central? Regards, Dave O'Brien From: Romanowski, John (OFT) Subject: Re: Performance problem Linux under Zvm If CP INDICATE QUEUES shows an En (like E3) in the 2nd column for one or more userids try CP QUERY SRM (write down response for reviewing ) and do this quick fix CP SET SRM STORBUF 300% 300% 300% This e-mail, including any attachments, may be confidential, privileged or otherwise legally protected. It is intended only for the addressee. If you received this e-mail in error or from someone who was not authorized to send it to you, do not disseminate, copy or otherwise use this e-mail or its attachments. Please notify the sender immediately by reply e-mail and delete the e-mail from your system. -Original Message- Subject: Performance problem Linux under Zvm Our shop is new to Zvm and Linux. We have a very small number of Linux users who are reporting significant response time problems. It almost seems as if each stops running for a period of time and is then re-dispatched. Is there a VM parameter that we might have taken the default on that needs tweaking? Any help or advice appreciated as this is a proof of concept endeavour and we would like not to turn off prospective users from the start. Thank you, Dave O'Brien National Institutes of Health
Re: Performance problem Linux under Zvm
I've put up my popular presentation configuring z/VM and Linux for Performance at velocitysoftware.com/present/config. This presentation looks at how to ensure your performance for Linux and z/VM is optimal, and provides the best practices. I'm still working on the notes. YOu will want to look at the SET SRM STORBUF setting towards the end of the presentation. O'Brien, David W. (NIH/CIT) [C] wrote: Our shop is new to Zvm and Linux. We have a very small number of Linux users who are reporting significant response time problems. It almost seems as if each stops running for a period of time and is then re-dispatched. Is there a VM parameter that we might have taken the default on that needs tweaking? Any help or advice appreciated as this is a proof of concept endeavour and we would like not to turn off prospective users from the start. Thank you, Dave O'Brien National Institutes of Health
Re: MONWRITE files
If you search google, using z/VM performance capacity planning, you should see velocitysoftware.com/whylps.html as the first link. This is the description of instrumentation requirements. I've also offered zMON (dirt cheap, by the way) as a real time monitor that will produce records for MICS or MXG. The benefit to this is that you have a monitor with a standard one minute granularity, but you write out the records that MICS or MXG want every 15 minutes. This reduces your disk requirements by usually more than factor of 100, and satisfies most basic needs for performance data AND capacity planning data (as well as operational alerts). Thomas Kern wrote: Because as someone pointed out before (Barton, I think), for performance monitoring, you want more event data and for capacity planning you need appropriate sample data. I think PerfTK could deal with more data than it needs for online real-time performance monitoring, while another process selectively collects from the same source a subset of that data for capac ity planning using that other z/OS product. /Tom Kern On Tue, 18 Mar 2008 11:01:07 -0400, Jim Bohnsack [EMAIL PROTECTED] wro te: Unless the objective is to have a program that will collect some of the records for one reporting function and another instance of the program collecting a different selection of the monitor records for a different reporting function, why not just limit the amount of records written to the monitor segment in the first place? Jim
Re: MONWRITE files
Wasn't going to answer, but Bill P goaded me to it. Is some monitor data worthless to be recorded? YES, LOTS OF IT for the MXG user. Is the shelf life short? YES for much of it. (Disks doing zero I/O but online?) Is MONWRITE writing garbage? Not all of it. To most users, much of the data is indeed garbage or at least something less than useful. We replaced MONWRITE in 1989 with intelligent operation that supports MXG the way MXG users like, supports MICS like MICS users like, and saved customers significant CPU on z/OS, disk space on both MVS (z/OS) and VM/XA (z/VM), and we can give Linux process, application and user data to both MXG and MICS. MONWRITE is old and in the way of progress. From our products page http://velocitysoftware.com/product.html; there is a link to our vendor interfaces such as MXG. We even try and support the IBM products, but there are few users. Alan Altmark wrote: On Wednesday, 03/12/2008 at 01:48 EDT, barton [EMAIL PROTECTED] wrote: Alan, look at what he's collecting. If you don't think that is miscollecting, you should take the class too. Are you telling me that some monitor data is worthless to be recorded? That is, that the shelf life is some of the data is so short that it has no value in a disk file? Or are you telling me that MONWRITE is writing garbage or double-writing or not writing or ...? If you're going to accuse MONWRITE of bad behavior, then I think you should tell me what you think it is doing that is bad. No FUD. Alan Altmark z/VM Development IBM Endicott