Constraint of zVM on number of Devices Support
Hi, I am new in zVM world. Can anyone let me know or tell me where can I find the zVM information of the number of devices can zVM support. Will it be any problems if I share the same IOCDS for zOS with zVM, where the number of devices defined in the IOCDS more than 16384. (The number of addresses for DASDs alone is 16384). FYI, we have 6 LPARs on the z9 machine and plan to run zVM on one of the LPARs. All the LPARs in z9 machine are sharing the same IOCDS. Thank you. Regards, Yee Fong Ooi 黄 宇 雄
Re: Constraint of zVM on number of Devices Support
I guess it is 64K (from to ) The infoirmation probably is in the z/VM General Information manual. But, why define all z/OS devices to z/VM? - It will consume some real storage in z/VM - You might by accident format a volume in z/VM that z/OS is using. - It costs some time to check them when z/VM is IPLed, and shut down - A CP Q DASD will yield a very long result I guess you would use address ranges for DASDs for use by z/VM, so define only these ranges to the z/VM partitions. 2008/10/31 Yee Fong Ooi [EMAIL PROTECTED]: Hi, I am new in zVM world. Can anyone let me know or tell me where can I find the zVM information of the number of devices can zVM support. Will it be any problems if I share the same IOCDS for zOS with zVM, where the number of devices defined in the IOCDS more than 16384. (The number of addresses for DASDs alone is 16384). FYI, we have 6 LPARs on the z9 machine and plan to run zVM on one of the LPARs. All the LPARs in z9 machine are sharing the same IOCDS. Thank you. Regards, Yee Fong Ooi 黄 宇 雄 -- Kris Buelens, IBM Belgium, VM customer support
Re: RACF ERROR
Thanks Alan! Thank You, Terry Martin Lockheed Martin - Information Technology z/OS z/VM Systems - Performance and Tuning Cell - 443 632-4191 Work - 410 786-0386 [EMAIL PROTECTED] -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Alan Altmark Sent: Thursday, October 30, 2008 12:28 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: RACF ERROR On Thursday, 10/30/2008 at 08:32 EDT, Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] wrote: One last question on this. This SMF CONTROL file change (1 to 0) does this setting normally mean that you have the RACFSMF guest in place and are using it for dynamic switching of the SMF data set and if you did not have the '1' there it would not invoke RACFSMF to do the switch if needed? In other words if you are using RACFSMF to do the switching would you need the '1' in the control file? You normally have RACFSMF in place, preventing SMF CONTROL from ever having a 1 in it in since both SMF disks never fill up at the same time). I mean, it's no guarantee, since there's no guarantee that RACFSMF will be successful in copying the SMF log to its own A-disk and archive. (If the copies fail, the originals aren't erased.) Alan Altmark z/VM Development IBM Endicott
Re: Share QDIO device between VM 1stlvl VM 2ndlvl
Kris, I tested what you told me and it works as expected. Just one thing, no need to code 192.20.2.182 in the HOME. It doesn't work. Just PROXYARP is needed ! Thanks Alain Le 29/10/08 8:54, « Kris Buelens » [EMAIL PROTECTED] a écrit : It can work: your first-level TCPIP needs to have PROXYARP in ASSORTEDPARMS and the IP address of the secondlevel TCPIP needs to be coded on the HOME of the first-level TCPIP: HOME 192.200.2.161 OSA82L 192.200.2.182 LNK00FD ; (I used such a setup when we got short on OSA ports on a 9672). If however if the only thing you need is 3270 access to the secondlevel VM, such a setup isn't required: - define some virtual 3270's in the config for the secondlevel guest (using SPECIAL or DEF GRAF) - ENABLE the addresses in the secondlevel - on the first-level VM logo, issue DIAL level2vm Even if you have this TCPIP setup, these virtual 3270s remain useful for cases where TCPIP in the secondlevel system would not start. 2008/10/28 David Kreuter [EMAIL PROTECTED] I'm not sure why you are using virtual CTCs. You should try to put the second level guest and the 1st level TCPIP on a vswitch. With the vswitch connecting via the osa the physical network both the 1st level vm tcpip and the tcpip 3rd level in the guest will have connectivity. Or, if you want to have two networks, keep the first level on the OSA, if that is what you are doing now, and then create a guest lan and give the 1st level and the 3rd level TCPIP on the guest interfaces to it. The first level will then act as a software router. David From: The IBM z/VM Operating System on behalf of Alain Benveniste Sent: Tue 10/28/2008 1:07 PM To: IBMVM@LISTSERV.UARK.EDU Subject: [IBMVM] Share QDIO device between VM 1stlvl VM 2ndlvl I would like to use a unique device, AA14, to connect 1stlvl with 2ndlvl.= I tried with CTC and it works inside the VMs. But I can't enter 2ndlvl wh= en I try to connect from the global network. I talked with the IP guy and told me it worked that way when it was coded= with LCS in place of QDIO. He has no idea how to do ... Config 1stlvl : DEVICE OSA82P OSD AA14 LINK OSA82L QDIOETHERNET OSA82P ; DEVICE CTC00FD CTC 101A LINK LNK00FD CTC 0 CTC00FD ; HOME 192.200.2.161 OSA82L ; GATEWAY ; (IP) Network First Link Max. Packet Subnet Subnet ; Address Hop Name Size (MTU) MaskValue ; --- --- --- --- -- 192.200.2.182 = LNK00FD 1500 HOST 192.200.2 = OSA82L1492 0 DEFAULTNET192.200.2.240 OSA82L1492 0 Config 2ndlvl : DEVICE CTC00FD CTC 00FD LINK LNK00FD CTC 1 CTC00FD ; HOME 192.200.2.182 LNK00FD ; GATEWAY ; (IP) Network First Link Max. Packet Subnet Subnet ; Address Hop Name Size (MTU) MaskValue ; --- --- --- --- -- 192.200.2 = LNK00FD 1492 0 DEFAULTNET192.200.2.240 LNK00FD 1492 0 ; What should I code to make it work ? Alain Benveniste -- Kris Buelens, IBM Belgium, VM customer support
CAVMEN Meeting on Thursday, November 13, 2008
The fourth quarter meeting of the Chicago Area VM (and Linux) Enthusiasts will be held on Thursday, November 13, 2008. We have had more speaker scheduling problems since my last note. The agenda has changed again. Please review it. At this point, we have only three main sessions confirmed. We will likely add a fourth, but I did not want to delay sending out notices any longer. If the agenda changes, we will send out another note. -- Meeting Location: This quarter's meeting will be held at the Hewitt Associates 'East Campus' located at 100 Half Day Road, in Lincolnshire, IL. We will meet in the Lower Level Conference Room in Building 98. If you have not attended a meeting at this location before, or you are not familiar with the area, http://cavmen.home.comcast.net/hewittb99.htmlClick here for additional information on directions, maps, lodging and dining. -- Attendance: We would like to request a count of expected attendees by the Monday before the meeting, so that we may plan appropriately for arranging the facilities, and for refreshments and lunch, should one of the vendors wish to provide them. If you are planning to attend, PLEASE send an E-Mail by that date to mailto:[EMAIL PROTECTED][EMAIL PROTECTED] with a subject line of Meeting Attendance. This is meant to be a facilities planning aid and should not be interpreted as a registration requirement. If you suddenly become available at the last minute, please feel free to attend even if you have not responded. Thank you in advance for your cooperation in this matter. -- Meeting Agenda: 9:00 AM z/VM Platform Update: Introducing z/VM 5.4 Advancing the Art of Server Virtualization. This presentation will highlight the new functions available with z/VM V5.4, IBM's advanced server virtualization solution for IBM System z. z/VM V5.4, generally available since September 12, 2008, offers enhanced virtualization capabilities which include: dynamic memory upgrade, flexible specialty engine configuration support, and virtual server scalability enhancements. Find out how z/VM V5.4 can help clients further leverage their System z infrastructure for improved business results. The speaker will be Rex Johnson of the IBM Corporation. 10:30 AMCoffee Break 11:00 AMPerformance Management Solutions in the z/Linux Environment IBM has greatly enhanced OMEGAMON's monitoring capability in the z/VM environment, in addition to providing full monitoring capability on z/Linux. In this session we will briefly discuss the current state of the product, touch on features added in the latest release, and provide a live demonstration of OMEGAMON. The demonstration will feature custom application views available using the Tivoli Enterprise Portal GUI, along with ability to automatically take action when problem conditions are found. The speaker will be Wayne Bucek of the IBM Corporation. 12:30 PMLunch Break 1:30 PM Administration and Vendor Announcements 1:45 PM Linux on z Update The speaker will present a brief update of recent developments in the Linux under z/VM area. The speaker will be Khaylen Kingsley of the IBM Corporation. 2:00 PM Installing a Novell SLES 10 Starter System without a Net(work) Multiple customers wishing to install Linux on System z have been frustrated by their own network firewall security rules preventing access from a z/VM mainframe to a CD-drive from which to load a Linux installation CD. Other customers have been frustrated by difficulties coordinating access to a separate Linux or UNIX systems elsewhere on their site network. Still others fear the 'newness' of the Linux environment, not knowing how to answer installation questions. This session will feature a live demonstration of a new Novell SLES 10 Starter System installation which requires nothing more than an existing z/VM system with a working 'FTPSERVE' server (supplied with z/VM), and a PC with internet access with which to download files from Novell. The speaker will be Mike Walter of Hewitt Associates. 3:15 PM Coffee Break and Prize Drawing 3:30 PM Free-for-All Members will attempt to answer any reasonable VM or hardware related questions. If you are having a problem and want to find out if others are experiencing it, or you are installing new hardware or software and want to find out what types of problems others have experienced, here is the place to find out. Members are encouraged to bring ideas for future presentations and speakers to this meeting. -- Please check the WEB site for Map and Directions: http://cavmen.home.comcast.net In addition, you will also find extensive information available on dining and lodging in the Hewitt Associates area. Additional information about the CAVMEN group, and other VM related items of interest are available on our web site. There is no charge
Re: I/O Overhead - z/VM versus VMWARE
David Kreuter wrote: Is non-insignifcant a mutated way of saying significant? Kind of. Litotes -- see http://en.wikipedia.org/wiki/Litotes ...phsiii (beating RSchuh to it, maybe!)
Re: I/O Overhead - z/VM versus VMWARE
The comparison is already complicated because of different terminology. For example what we consider the overhead of I/O Is that in CPU resources, memory or I/O resources? A Linux system for example use a cache to avoid I/O, but that means there is memory overhead for I/O. And when the page cache is very large, there is also CPU overhead in managing that cache. Depending on the cache hit ratio and the disk response time, the overhead may result in faster response. And who would measure that on VMware? When talking to people from the other side it seems that I/O overhead is sometimes the measured throughput of an (assumed) I/O bound application, running either on VMware or native. To me such an interpretation has more questions than answers. But VM people tend to look at the T/V ratio of a virtual machine running such an application. That's not entirely fair because it also includes the cost of security, performance instrumentation, error recovery, etc. If a virtual machine wants to do I/O it may require paging to make room for the new data - is that I/O overhead? In general, I don't think that the difference in I/O performance is a motivation to run in LPAR rather than in a virtual machine. Before 5.2 we had the 2G issues that affected many installations. z/VM 5.3 addressed some more of these issues. One factor I do know about is MDC: many new workload does not exploit it as much as we used to do, so it sometimes good to consider not spending the resources on that. One of the factors that influences virtual machine I/O is the extra latency that comes with running multiple virtual machines. While this does not impact the overall throughput of the configuration, it does affect the maximum single thread throughput. That's what makes benchmarking complicated in this environment. QDIO (as used for OSA devices and FCP) is meant to address this latency issue. Under proper conditions this allows the channel to drive the I/O operations without waiting for the virtual machine to be dispatched and issue the next SSCH. This does have its price though, the CPU usage is higher than with ECKD I/O, so if you're CPU constrained the I/O might become slower. Rob -- Rob van der Heij Velocity Software http://www.velocitysoftware.com/
Re: I/O Overhead - z/VM versus VMWARE
In the Z10-BC web announcement there was a comarison of Z/VM and VMware running LINUX guests. I have the PDF presentation if you want it I can send it to you.. Or it is probably not too difficult to find on IBM's web site. -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] Behalf Of Alan Ackerman Sent: Thursday, October 30, 2008 6:25 PM To: IBMVM@LISTSERV.UARK.EDU Subject: I/O Overhead - z/VM versus VMWARE I got asked: “Does z/VM impose non-insignificant overhead? Is it similar to VMware,= in which virtual I/O imposes significant overhead, but most processor and = memory access runs at close to native physical speed?” I don’t know anything about VMWARE so I could not answer the question. = I know that CCW Translation in VM costs significant cycles. I think FCP disks dedicated DASD fullpack minidisks small minidisks= . I would HOPE that the zSeries, with so much of virtualization built into = the hardware, would have lower costs than VMWARE, but I don’t really kn= ow. Any takers? Are there any web sites that give performance comparisons VM versus VMWAR= E? Alan Ackerman= Alan (dot) Ackerman (at) Bank of America (dot) com
Re: I/O Overhead - z/VM versus VMWARE
I think you need to compare Linux in a z/VM LPAR versus Linux Native, and determine if one has more i/o overhead than the other. I think the answer is going to be 'minimal'. As someone said, and I have observed, my CP%CPU runs at 2-3%. As for MDC, I've been curious about that lately. About a week ago, I turned off mdc for a highly active volume, and it seemed to me that resp increased rapidly and markedly. I quickly turned it back on. So I tried just now, for fun. I took the highest activity volume in a large system, and turned MDC cache off. Here's the info at 5min intervals: I/O RateResp 127 3.0 MDC ON 86.5 6.5MDC OFF 69.6 9.2 102 4.9 122 3.3 93 3.5 11.7 12.3 At that point its activity seemed to decrease. So I guess my test is shot. :) MA On Fri, Oct 31, 2008 at 8:49 AM, Rob van der Heij [EMAIL PROTECTED] wrote: The comparison is already complicated because of different terminology. For example what we consider the overhead of I/O Is that in CPU resources, memory or I/O resources? A Linux system for example use a cache to avoid I/O, but that means there is memory overhead for I/O. And when the page cache is very large, there is also CPU overhead in managing that cache. Depending on the cache hit ratio and the disk response time, the overhead may result in faster response. And who would measure that on VMware? When talking to people from the other side it seems that I/O overhead is sometimes the measured throughput of an (assumed) I/O bound application, running either on VMware or native. To me such an interpretation has more questions than answers. But VM people tend to look at the T/V ratio of a virtual machine running such an application. That's not entirely fair because it also includes the cost of security, performance instrumentation, error recovery, etc. If a virtual machine wants to do I/O it may require paging to make room for the new data - is that I/O overhead? In general, I don't think that the difference in I/O performance is a motivation to run in LPAR rather than in a virtual machine. Before 5.2 we had the 2G issues that affected many installations. z/VM 5.3 addressed some more of these issues. One factor I do know about is MDC: many new workload does not exploit it as much as we used to do, so it sometimes good to consider not spending the resources on that. One of the factors that influences virtual machine I/O is the extra latency that comes with running multiple virtual machines. While this does not impact the overall throughput of the configuration, it does affect the maximum single thread throughput. That's what makes benchmarking complicated in this environment. QDIO (as used for OSA devices and FCP) is meant to address this latency issue. Under proper conditions this allows the channel to drive the I/O operations without waiting for the virtual machine to be dispatched and issue the next SSCH. This does have its price though, the CPU usage is higher than with ECKD I/O, so if you're CPU constrained the I/O might become slower. Rob -- Rob van der Heij Velocity Software http://www.velocitysoftware.com/
Re: I/O Overhead - z/VM versus VMWARE
On Fri, Oct 31, 2008 at 2:30 PM, Mary Anne Matyaz [EMAIL PROTECTED] wrote: As for MDC, I've been curious about that lately. About a week ago, I turned off mdc for a highly active volume, and it seemed to me that resp increased rapidly and markedly. It's good that you try and measure rather than rely on hearsay or guts feeling. The nasty part is that I/O measurement is complicated (that's why I have been postponing such research to the point where I have more time to spend on it). When you set MDC OFF for a virtual machine or virtual device, it avoids further inserts but anything in MDC will remain there and gets used on a read (unless you also purge it). I recall from some experiments in the past that system-wide MDC OFF was the only thing that really made a difference when the workload did not take advantage of it. Further, the actual I/O performed by CP is may be different with MDC enabled. When Linux wants a series of blocks and one is found in MDC, the rest is read through 2 I/O operations by CP. The consequence is that the size of the average I/O goes down, so the I/O response time (per I/O operation) gets lower. Whether that makes the throughput of the application higher with the same factor is not obvious. Another gotcha is that MDC takes the low-hanging fruit. So when you enable MDC the remaining I/O to the DASD subsystem are less likely to cache than when you would do all I/O. :anecdote type=sad. Long ago, we replaced some 3390's with a RAID based subsystem. The vendor had promised a certain cache hit ratio internally in the subsystem. When that was not met for VM devices, we were told to disable MDC because it interfered with the DASD subsystem. Clearly, when VM MDC already avoided the ones that were easy to cache, the remaining work for the DASD was harder. But going to the subsystem cache is still slower than taking it out of MDC, so the application throughput got worse... :eanecdote. Rob -- Rob van der Heij Velocity Software http://velocitysoftware.com/
Re: I/O Overhead - z/VM versus VMWARE
Similar story at my former customer: At a certain time, the VM systems got a DASD subsystem that MVS no longer needed as it was replaced by a more modern one. These MVS guys wanted to see our IO responsetime, which was worse than on MVS. Conclusion (from the MVS guys): VM isn't as good as MVS. I had to repeat over and over again that VM's MDC avoided the IO's that cache best. To my surprize: when we started sharing a DS8000 with z/OS, the I/O response times of z/VM and z/OS became close. 2008/10/31 Rob van der Heij [EMAIL PROTECTED]: On Fri, Oct 31, 2008 at 2:30 PM, Mary Anne Matyaz [EMAIL PROTECTED] wrote: As for MDC, I've been curious about that lately. About a week ago, I turned off mdc for a highly active volume, and it seemed to me that resp increased rapidly and markedly. It's good that you try and measure rather than rely on hearsay or guts feeling. The nasty part is that I/O measurement is complicated (that's why I have been postponing such research to the point where I have more time to spend on it). When you set MDC OFF for a virtual machine or virtual device, it avoids further inserts but anything in MDC will remain there and gets used on a read (unless you also purge it). I recall from some experiments in the past that system-wide MDC OFF was the only thing that really made a difference when the workload did not take advantage of it. Further, the actual I/O performed by CP is may be different with MDC enabled. When Linux wants a series of blocks and one is found in MDC, the rest is read through 2 I/O operations by CP. The consequence is that the size of the average I/O goes down, so the I/O response time (per I/O operation) gets lower. Whether that makes the throughput of the application higher with the same factor is not obvious. Another gotcha is that MDC takes the low-hanging fruit. So when you enable MDC the remaining I/O to the DASD subsystem are less likely to cache than when you would do all I/O. :anecdote type=sad. Long ago, we replaced some 3390's with a RAID based subsystem. The vendor had promised a certain cache hit ratio internally in the subsystem. When that was not met for VM devices, we were told to disable MDC because it interfered with the DASD subsystem. Clearly, when VM MDC already avoided the ones that were easy to cache, the remaining work for the DASD was harder. But going to the subsystem cache is still slower than taking it out of MDC, so the application throughput got worse... :eanecdote. Rob -- Rob van der Heij Velocity Software http://velocitysoftware.com/ -- Kris Buelens, IBM Belgium, VM customer support
Re: Constraint of zVM on number of Devices Support
I thought VP Q DASD will only show the CP and user defined owned volume list in the system config ..
Newbie On this list...Question RE: LISTSERV Delivery
Hi, Everyone... Quick question - I just want to get a daily digest...which I do...but I am also receiving individual emails, which I don't want. I don't know why I am getting both, since it appears in the LISTSERV settings that it can be one way or the other for one particular email address. Can someone help me correct this? Thank you! Andy Cooper | A c x i o m CDC | z/OS and z/VM Software Services Acxiom Software Services 312-985-3465 office | 312-287-2533 cell 555 W. Adams Street | Chicago, IL 60661 | USA [EMAIL PROTECTED] blocked::mailto:[EMAIL PROTECTED] __ ACXIOM(r) WE MAKE INFORMATION INTELLIGENTTM P Save a tree. Don't print this e-mail unless it's really necessary.
Re: Newbie On this list...Question RE: LISTSERV Delivery
On: Fri, Oct 31, 2008 at 09:50:27AM -0500,Cooper, Andy Wrote: } Quick question - I just want to get a daily digest...which I do...but I } am also receiving individual emails, which I don't want. I don't know } why I am getting both, since it appears in the LISTSERV settings that it } can be one way or the other for one particular email address. Andy, While that could be a Listserv glitch, I doubt it. The usual cause of this is that you are subscribed as 2 addresses. Did your email change at some point. -- Rich Greenberg N Ft Myers, FL, USA richgr atsign panix.com + 1 239 543 1353 Eastern time. N6LRT I speak for myself my dogs only.VM'er since CP-67 Canines:Val, Red, Shasta Casey (RIP), Red Zero, Siberians Owner:Chinook-L Retired at the beach Asst Owner:Sibernet-L
Re: I/O Overhead - z/VM versus VMWARE
On Thursday, 10/30/2008 at 10:29 EDT, David Kreuter [EMAIL PROTECTED] wrote: Is non-insignifcant a mutated way of saying significant? The i/o code in vm is like much of CP highly optimized. Overhead has been reduced greatly since the XA introduction of SIE emulation. VM has low overhead due to: 1. avoidance. Let SIE handle it. 2. highly optimized code paths. While I will grant you the optimization point, let's not get too carried away. In an LPAR, SIE handles guest I/O only for dedicated OSA and FCP adapters. All other I/O is virtualized by CP. SIE *does* handle CP's I/O! The emphasis on the use of a virtual switch rather than dedicated OSAs leaves us with primarily FCP adapters for SCSI. But with all that said, as others have pointed out, the word overhead has no meaning. Yes, there is overhead and sometimes, yes, it can be not insignificant. The question is whether the applications are meeting their SLAs and whether the IT provider is meeting its expense goals. Can I get acceptable application performance at a cost I can afford? As was mentioned, you may have more overhead handling an I/O request, but if you can satisfy it from MDC, it was time well-spent. Assuming, of course, you've got the CPU available to handle the I/O request! Alan Altmark z/VM Development IBM Endicott
Re: Newbie On this list...Question RE: LISTSERV Delivery
No, Rich, it didn't. I even totally unsubscribed and re-subscribed as daily digest only - and, obviously, I am still getting the single emails. Andy Cooper | A c x i o m CDC | z/OS and z/VM Software Services Acxiom Software Services 312-985-3465 office | 312-287-2533 cell 555 W. Adams Street | Chicago, IL 60661 | USA [EMAIL PROTECTED] __ ACXIOM(r) WE MAKE INFORMATION INTELLIGENTTM P Save a tree. Don't print this e-mail unless it's really necessary. -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Rich Greenberg Sent: Friday, October 31, 2008 10:36 AM To: IBMVM@listserv.uark.edu Subject: Re: Newbie On this list...Question RE: LISTSERV Delivery On: Fri, Oct 31, 2008 at 09:50:27AM -0500,Cooper, Andy Wrote: } Quick question - I just want to get a daily digest...which I do...but I } am also receiving individual emails, which I don't want. I don't know } why I am getting both, since it appears in the LISTSERV settings that it } can be one way or the other for one particular email address. Andy, While that could be a Listserv glitch, I doubt it. The usual cause of this is that you are subscribed as 2 addresses. Did your email change at some point. -- Rich Greenberg N Ft Myers, FL, USA richgr atsign panix.com + 1 239 543 1353 Eastern time. N6LRT I speak for myself my dogs only.VM'er since CP-67 Canines:Val, Red, Shasta Casey (RIP), Red Zero, Siberians Owner:Chinook-L Retired at the beach Asst Owner:Sibernet-L
Re: Constraint of zVM on number of Devices Support
We have more than that in our disaster recovery environment and we don't do anything to segregate VM from z/OS there. So, VM is ok with that - it does take longer to IPL though. For production, we separate out by LCU's (VM owns its own) and we restrict the gen so that z/VM can't see the z/OS devices (although z/OS can see z/VM's because it is required for Hyperswap to work). You could also provide segregation in the system config with what is online at IPL and what is accepted. (sorry for the HTML format - for some reason I can't change it on this particular email - weird). 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 Yee Fong Ooi Sent: Friday, October 31, 2008 5:02 AM To: IBMVM@LISTSERV.UARK.EDU Subject: [IBMVM] Constraint of zVM on number of Devices Support Hi, I am new in zVM world. Can anyone let me know or tell me where can I find the zVM information of the number of devices can zVM support. Will it be any problems if I share the same IOCDS for zOS with zVM, where the number of devices defined in the IOCDS more than 16384. (The number of addresses for DASDs alone is 16384). FYI, we have 6 LPARs on the z9 machine and plan to run zVM on one of the LPARs. All the LPARs in z9 machine are sharing the same IOCDS. Thank you.
Re: I/O Overhead - z/VM versus VMWARE
I agree. My point remains that most of the virtual machines' use of the instruction set is handled in SIE. And, while not getting carried away, great strides of been made in CP I/O handling, like fast path, etc. And avoidance with using MDC, shared segments, etc. also contributes to our sunny io disposition. David From: The IBM z/VM Operating System on behalf of Alan Altmark Sent: Fri 10/31/2008 11:37 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: [IBMVM] I/O Overhead - z/VM versus VMWARE On Thursday, 10/30/2008 at 10:29 EDT, David Kreuter [EMAIL PROTECTED] wrote: Is non-insignifcant a mutated way of saying significant? The i/o code in vm is like much of CP highly optimized. Overhead has been reduced greatly since the XA introduction of SIE emulation. VM has low overhead due to: 1. avoidance. Let SIE handle it. 2. highly optimized code paths. While I will grant you the optimization point, let's not get too carried away. In an LPAR, SIE handles guest I/O only for dedicated OSA and FCP adapters. All other I/O is virtualized by CP. SIE *does* handle CP's I/O! The emphasis on the use of a virtual switch rather than dedicated OSAs leaves us with primarily FCP adapters for SCSI. But with all that said, as others have pointed out, the word overhead has no meaning. Yes, there is overhead and sometimes, yes, it can be not insignificant. The question is whether the applications are meeting their SLAs and whether the IT provider is meeting its expense goals. Can I get acceptable application performance at a cost I can afford? As was mentioned, you may have more overhead handling an I/O request, but if you can satisfy it from MDC, it was time well-spent. Assuming, of course, you've got the CPU available to handle the I/O request! Alan Altmark z/VM Development IBM Endicott
Re: VM Newbie alert - Install from FTP server
Mark, Sorry for being late after vacation. I'd recommend Colin's suggestion in either case: There is no better playground for the people you mentioned than a 2nd level VM!!! If you have the disk resources you even can give each trainee its own 2nd level z/VM system where he/she can test what ever he/she wants without disturbing anybody else. If you want to isolate this as much as possible - the better. Just bring a copy of any other z/VM system to your free partition, define the virtual machines to host the 2nd level test systems, and the games are open. Wolfgang Software AG - Sitz/Registered office: Uhlandstra?e 12, 64297 Darmstadt, Germany, - Registergericht/Commercial register: Darmstadt HRB 1562 - Vorstand/ Management Board: Karl-Heinz Streibich (Vorsitzender/Chairman), David Broadbent, Mark Edwards, Holger Friedrich, Dr. Peter Kurpick, Arnd Zinnhardt; - Aufsichtsratsvorsitzender/ Chairman of the Supervisory Board: Frank F. Beelitz - http://www.softwareag.com -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Mark Jacobs Sent: Wednesday, October 08, 2008 2:26 PM To: ibmvm@listserv.uark.edu Subject: Re: VM Newbie alert - Install from FTP server Colin Allinson wrote: *Mark Jacobs [EMAIL PROTECTED]* wrote :- Actually we will be attempting to install zVM 5.4 on a new lpar so we will be using the integrated 3270 console on the HMC to preform the initial install. For additional fun we will be attempting to install everything directly to FCP DASD not ECKD. OK - that is different to what we do so my cookbook will not help you. I have just read the Summary card again and the instructions for 1st level from an FTP server are there and seem fairly clear - although I can't verify if they are correct. Just out of interest - if you have another VM LPAR, why did you decide to install 1st level rather than build it first at second level on another LPAR and then IPL it on you target LPAR when you know it is good? * Colin Allinson** * Amadeus Data Processing GmbH This lpar is going to be used as a system programmer sandbox where lots of people who can hardly spell zVM will be learning to install/administrate it. We want it to be as isolated from production as possible. -- Mark Jacobs Time Customer Service Tampa, FL Today, we celebrate the first glorious anniversary of the Information Purification Directives. We have created, for the first time in all history, a garden of pure ideology. Where each worker may bloom secure from the pests of contradictory and confusing truths. Our Unification of Thoughts is more powerful a weapon than any fleet or army on earth. We are one people, with one will, one resolve, one cause. Our enemies shall talk themselves to death and we will bury them with their own confusion. We shall prevail! Apple's television commercial - Super Bowl - 1984
Re: Constraint of zVM on number of Devices Support
Here, many of the devices (approximately 5000) are in the LPAR profile but excluded from VM's configuration by being included in a Not_Accepted list in SYSTEM CONFIG. That way, they do not cause the RDEVs to be built, but can be added via command if we ever need to access them from VM. Regards, Richard Schuh -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Kris Buelens Sent: Friday, October 31, 2008 5:10 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Constraint of zVM on number of Devices Support I guess it is 64K (from to ) The infoirmation probably is in the z/VM General Information manual. But, why define all z/OS devices to z/VM? - It will consume some real storage in z/VM - You might by accident format a volume in z/VM that z/OS is using. - It costs some time to check them when z/VM is IPLed, and shut down - A CP Q DASD will yield a very long result I guess you would use address ranges for DASDs for use by z/VM, so define only these ranges to the z/VM partitions. 2008/10/31 Yee Fong Ooi [EMAIL PROTECTED]: Hi, I am new in zVM world. Can anyone let me know or tell me where can I find the zVM information of the number of devices can zVM support. Will it be any problems if I share the same IOCDS for zOS with zVM, where the number of devices defined in the IOCDS more than 16384. (The number of addresses for DASDs alone is 16384). FYI, we have 6 LPARs on the z9 machine and plan to run zVM on one of the LPARs. All the LPARs in z9 machine are sharing the same IOCDS. Thank you. Regards, Yee Fong Ooi 黄 宇 雄 -- Kris Buelens, IBM Belgium, VM customer support
Re: Reliability of SFS?
On Thu, Oct 30, 2008 at 2:38 PM, Alan Altmark [EMAIL PROTECTED] wrote: It's not just about intermediate nodes. It's also about the node that runs the SFS server. While it will impact those who need the data before you restart the server elsewhere, dropping the directory also impacts those who want it later. My major concern was about SFS control file backup, until I recently learned that you can also make it point to a directory rather than the file mode of a previously accessed directory. So that is gone now... I appreciate the auto-release issue, but I don't see us changing that behavior. It would be a Big Deal to redesign the SFS client. For myself, I'd probably write a nucleus extension that periodically tries to re-access my preferred filemodes if they aren't accessed. That idea does not fly. When my program needs the directory right now, it is of no help that it will be re-accessed in 5 seconds. The programming model around mini disk and file mode is that the disk is there from access until you release it yourself (or very bad things have happened that make a virtual machine question whether it is worth living). I don't want to program my code catch such an error on each I/O (if you even can). Rob
Re: I/O Overhead - z/VM versus VMWARE
Hi Thomas, Could you send me a copy of this pdf presentation? Thanks P On Fri, Oct 31, 2008 at 6:28 AM, Huegel, Thomas [EMAIL PROTECTED] wrote: In the Z10-BC web announcement there was a comarison of Z/VM and VMware running LINUX guests. I have the PDF presentation if you want it I can send it to you.. Or it is probably not too difficult to find on IBM's web site. -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] Behalf Of Alan Ackerman Sent: Thursday, October 30, 2008 6:25 PM To: IBMVM@LISTSERV.UARK.EDU Subject: I/O Overhead - z/VM versus VMWARE I got asked: Does z/VM impose non-insignificant overhead? Is it similar to VMware,= in which virtual I/O imposes significant overhead, but most processor and = memory access runs at close to native physical speed? I don't know anything about VMWARE so I could not answer the question. = I know that CCW Translation in VM costs significant cycles. I think FCP disks dedicated DASD fullpack minidisks small minidisks= . I would HOPE that the zSeries, with so much of virtualization built into = the hardware, would have lower costs than VMWARE, but I don't really kn= ow. Any takers? Are there any web sites that give performance comparisons VM versus VMWAR= E? Alan Ackerman= Alan (dot) Ackerman (at) Bank of America (dot) com
Re: I/O Overhead - z/VM versus VMWARE
Too busy to look up the reference. Besides, you used the word on this list first, so it is only right. :-) Regards, Richard Schuh -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Phil Smith III Sent: Friday, October 31, 2008 5:35 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: I/O Overhead - z/VM versus VMWARE David Kreuter wrote: Is non-insignifcant a mutated way of saying significant? Kind of. Litotes -- see http://en.wikipedia.org/wiki/Litotes ...phsiii (beating RSchuh to it, maybe!)
Re: Recycle yourself
Put the Linux machines under AUDITORs control. If you want something more immediate, you can code a process in PROP (or any of the other console managers). (i.e. look for LINUX70 LOGGED OFF and execute a XAUTOLOG) Tom Duerbusch THD Consulting Scott Rohling [EMAIL PROTECTED] 10/29/2008 8:37 PM Right -- SIGNAL I know about.. but you can only SIGNAL SHUTDOWN .. How about a SIGNAL SHUTDOWN -R ;-) Thanks - Scott On Wed, Oct 29, 2008 at 6:50 PM, Nick Laflamme [EMAIL PROTECTED] wrote: On Oct 29, 2008, at 7:17 PM, Scott Rohling wrote: Is there a 'native' way to have your guest brought down and autologged? I suppose I'm looking for a CP command which instead of allowing the guest to say.. IPL - actually signals it off If this were You Bet Your Life, you'd win the prize, for SIGNAL is the command you're looking for. It only works if the guest registers with CP to receive signals, but it would suit your purposes. Romney White once published an example of CMS code that lets you load a CMS nucleus extension to catch a signal. Linux has supported it for years. And, of course, there's the CP command, SIGNAL, to manually initiate a SIGNAL to a specific guest. Nothing tells the guest to re-start itself, so perhaps we'd want a second signal besides, SHUTDOWN to differentiate between die, and die and come back. But, you're in the right neighborhood.
Re: Reliability of SFS?
On Friday, 10/31/2008 at 12:19 EDT, Rob van der Heij [EMAIL PROTECTED] wrote: It's not just about intermediate nodes. It's also about the node that runs the SFS server. While it will impact those who need the data before you restart the server elsewhere, dropping the directory also impacts those who want it later. I appreciate the auto-release issue, but I don't see us changing that behavior. It would be a Big Deal to redesign the SFS client. For myself, I'd probably write a nucleus extension that periodically tries to re-access my preferred filemodes if they aren't accessed. That idea does not fly. When my program needs the directory right now, it is of no help that it will be re-accessed in 5 seconds. The programming model around mini disk and file mode is that the disk is there from access until you release it yourself (or very bad things have happened that make a virtual machine question whether it is worth living). I don't want to program my code catch such an error on each I/O (if you even can). I don't understand, Rob. Every time a program writes, it must check EVERY call it makes to ensure it worked. It can't just blindly continue. In fact, if the connection between client and server is broken, all changes to any open files are backed out, so the program must be restarted anyway. (Workunits, you know!) And if a minidisk is DETACHed and LINKed, CMS doesn't automatically ACCESS it for you. I don't know why you expect SFS to be different in this respect. Availability of SFS servers in a clustered environment is provided by automation. If FPOOL1 is normally running on SYSTEM1 and SYSTEM1 dies, automation brings FPOOL1 up on SYSTEM2 (shared MR mdisks). CSE, if used, will prevent FPOOL1 from logging onto both System1 and System2 at the same time. If CSE isn't used, then the cross-system links will protect the minidisks and CP will ensure that there's only one owner of the filepool (resource) name. Alan Altmark z/VM Development IBM Endicott
Command for find out % of SMF utilization
Hi Is there a command that I can use to tell me the % used of the SMF data sets? Thank You, Terry Martin Lockheed Martin - Information Technology z/OS z/VM Systems - Performance and Tuning Cell - 443 632-4191 Work - 410 786-0386 [EMAIL PROTECTED]
Re: Newbie On this list...Question RE: LISTSERV Delivery
On: Fri, Oct 31, 2008 at 10:38:22AM -0500,Cooper, Andy Wrote: } No, Rich, it didn't. I even totally unsubscribed and re-subscribed as } daily digest only - and, obviously, I am still getting the single } emails. It appears that unsubscribing from the old address didn't take then. If you can still send mail From: the old address, send the listserv a Query IBMVM from the old address, and if that response indicates you are still subscribed from there, unsubscribe from that address again. Beyond that, you will have to get Dan involved. Reach him at: List-Owner: mailto:[EMAIL PROTECTED] -- Rich Greenberg N Ft Myers, FL, USA richgr atsign panix.com + 1 239 543 1353 Eastern time. N6LRT I speak for myself my dogs only.VM'er since CP-67 Canines:Val, Red, Shasta Casey (RIP), Red Zero, Siberians Owner:Chinook-L Retired at the beach Asst Owner:Sibernet-L
Re: Command for find out % of SMF utilization
The bare minimum: VMLINK RACFVM 301 * Z QUERY DISK Z VMLINK RACFVM 301 DETACH 2008/10/31 Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED]: Hi Is there a command that I can use to tell me the % used of the SMF data sets? Thank You, Terry Martin Lockheed Martin - Information Technology z/OS z/VM Systems - Performance and Tuning Cell - 443 632-4191 Work - 410 786-0386 [EMAIL PROTECTED] -- Kris Buelens, IBM Belgium, VM customer support
Re: Command for find out % of SMF utilization
vmlink racfvm 301 (i q disk .fm Same for 302 Scott On Fri, Oct 31, 2008 at 11:08 AM, Kris Buelens [EMAIL PROTECTED]wrote: The bare minimum: VMLINK RACFVM 301 * Z QUERY DISK Z VMLINK RACFVM 301 DETACH 2008/10/31 Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED]: Hi Is there a command that I can use to tell me the % used of the SMF data sets? Thank You, Terry Martin Lockheed Martin - Information Technology z/OS z/VM Systems - Performance and Tuning Cell - 443 632-4191 Work - 410 786-0386 [EMAIL PROTECTED] -- Kris Buelens, IBM Belgium, VM customer support
Re: I/O Overhead - z/VM versus VMWARE
Phil, Pretty not too bad post! ;-) Mike Walter Phil Smith III [EMAIL PROTECTED] Sent by: The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU 10/31/2008 07:34 AM Please respond to The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU To IBMVM@LISTSERV.UARK.EDU cc Subject Re: I/O Overhead - z/VM versus VMWARE David Kreuter wrote: Is non-insignifcant a mutated way of saying significant? Kind of. Litotes -- see http://en.wikipedia.org/wiki/Litotes ...phsiii (beating RSchuh to it, maybe!) 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.
Re: I/O Overhead - z/VM versus VMWARE
Hello Thomas, I would also appreciate a copy of this PDF. TIA HITACHI DATA SYSTEMS Raymond E. Noal Senior Technical Engineer Office: (408) 970 - 7978 From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Patrick Leigh Sent: Friday, October 31, 2008 9:23 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: I/O Overhead - z/VM versus VMWARE Hi Thomas, Could you send me a copy of this pdf presentation? Thanks P On Fri, Oct 31, 2008 at 6:28 AM, Huegel, Thomas [EMAIL PROTECTED] wrote: In the Z10-BC web announcement there was a comarison of Z/VM and VMware running LINUX guests. I have the PDF presentation if you want it I can send it to you.. Or it is probably not too difficult to find on IBM's web site. -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] Behalf Of Alan Ackerman Sent: Thursday, October 30, 2008 6:25 PM To: IBMVM@LISTSERV.UARK.EDU Subject: I/O Overhead - z/VM versus VMWARE I got asked: Does z/VM impose non-insignificant overhead? Is it similar to VMware,= in which virtual I/O imposes significant overhead, but most processor and = memory access runs at close to native physical speed? I don't know anything about VMWARE so I could not answer the question. = I know that CCW Translation in VM costs significant cycles. I think FCP disks dedicated DASD fullpack minidisks small minidisks= . I would HOPE that the zSeries, with so much of virtualization built into = the hardware, would have lower costs than VMWARE, but I don't really kn= ow. Any takers? Are there any web sites that give performance comparisons VM versus VMWAR= E? Alan Ackerman= Alan (dot) Ackerman (at) Bank of America (dot) com
Re: I/O Overhead - z/VM versus VMWARE
Raymond.can you send me a copy as well.Please..thanks Danny Padilla (623) 255 1553 _ From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Raymond Noal Sent: Friday, October 31, 2008 11:26 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: I/O Overhead - z/VM versus VMWARE Hello Thomas, I would also appreciate a copy of this PDF. TIA HITACHI DATA SYSTEMS Raymond E. Noal Senior Technical Engineer Office: (408) 970 - 7978 _ From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Patrick Leigh Sent: Friday, October 31, 2008 9:23 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: I/O Overhead - z/VM versus VMWARE Hi Thomas, Could you send me a copy of this pdf presentation? Thanks P On Fri, Oct 31, 2008 at 6:28 AM, Huegel, Thomas [EMAIL PROTECTED] wrote: In the Z10-BC web announcement there was a comarison of Z/VM and VMware running LINUX guests. I have the PDF presentation if you want it I can send it to you.. Or it is probably not too difficult to find on IBM's web site. -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] Behalf Of Alan Ackerman Sent: Thursday, October 30, 2008 6:25 PM To: IBMVM@LISTSERV.UARK.EDU Subject: I/O Overhead - z/VM versus VMWARE I got asked: Does z/VM impose non-insignificant overhead? Is it similar to VMware,= in which virtual I/O imposes significant overhead, but most processor and = memory access runs at close to native physical speed? I don't know anything about VMWARE so I could not answer the question. = I know that CCW Translation in VM costs significant cycles. I think FCP disks dedicated DASD fullpack minidisks small minidisks= . I would HOPE that the zSeries, with so much of virtualization built into = the hardware, would have lower costs than VMWARE, but I don't really kn= ow. Any takers? Are there any web sites that give performance comparisons VM versus VMWAR= E? Alan Ackerman= Alan (dot) Ackerman (at) Bank of America (dot) com
New version of IPGATE....
Thanks to the efforts of Perry Ruiter, there is now a new version of IPGATE available. This version contains updates to IPGATE by Kris Buelens to support CMS Alternate User ID and by Perry Ruiter to fix connection timeout errors and to allow IPGATE traffic to flow over links that are not on the default home address. It can be found here: http://www.vsoft-software.com/ipgate.vmarc Enjoy. -- DJ V/Soft z/VM and mainframe Linux expertise, training, consulting, and software development www.vsoft-software.com
Re: Command for find out % of SMF utilization
Thanks all for the help! Thank You, Terry Martin Lockheed Martin - Information Technology z/OS z/VM Systems - Performance and Tuning Cell - 443 632-4191 Work - 410 786-0386 [EMAIL PROTECTED] From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Scott Rohling Sent: Friday, October 31, 2008 1:16 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Command for find out % of SMF utilization vmlink racfvm 301 (i q disk .fm Same for 302 Scott On Fri, Oct 31, 2008 at 11:08 AM, Kris Buelens [EMAIL PROTECTED] wrote: The bare minimum: VMLINK RACFVM 301 * Z QUERY DISK Z VMLINK RACFVM 301 DETACH 2008/10/31 Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED]: Hi Is there a command that I can use to tell me the % used of the SMF data sets? Thank You, Terry Martin Lockheed Martin - Information Technology z/OS z/VM Systems - Performance and Tuning Cell - 443 632-4191 Work - 410 786-0386 [EMAIL PROTECTED] -- Kris Buelens, IBM Belgium, VM customer support
Re: Newbie On this list...Question RE: LISTSERV Delivery
Hi Andy, Listserv cannot produce both a daily digest and individual distributions from the same subscriptions, so you have 2 subscriptions. If you cannot determine the subscription address of the subscription you don't want, you need the help of a list owner. For any Listserv list, the address of the list owner is [EMAIL PROTECTED], or in this case: [EMAIL PROTECTED] The problem could be as simple as a subscription address different from, but being read on your main/expected e-mail account, or could be something as subtle as one subscription being for acooper and the other ACooper (usually case doesn't matter, but sometimes it does). cheers, wayne (a list owner, but not an IBMVM list owner!) Cooper, Andy wrote, in part, on 2008-10-31 10:50 AM: Quick question – I just want to get a daily digest…which I do…but I am also receiving individual emails, which I don’t want. I don’t know why I am getting both, since it appears in the LISTSERV settings that it can be one way or the other for one particular email address.
Re: I/O Overhead - z/VM versus VMWARE
Danny, Here you go - HITACHI DATA SYSTEMS Raymond E. Noal Senior Technical Engineer Office: (408) 970 - 7978 From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Danny Padilla Sent: Friday, October 31, 2008 12:50 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: I/O Overhead - z/VM versus VMWARE Importance: High Raymond...can you send me a copy as well...Pleasethanks Danny Padilla (623) 255 1553 From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Raymond Noal Sent: Friday, October 31, 2008 11:26 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: I/O Overhead - z/VM versus VMWARE Hello Thomas, I would also appreciate a copy of this PDF. TIA HITACHI DATA SYSTEMS Raymond E. Noal Senior Technical Engineer Office: (408) 970 - 7978 From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Patrick Leigh Sent: Friday, October 31, 2008 9:23 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: I/O Overhead - z/VM versus VMWARE Hi Thomas, Could you send me a copy of this pdf presentation? Thanks P On Fri, Oct 31, 2008 at 6:28 AM, Huegel, Thomas [EMAIL PROTECTED] wrote: In the Z10-BC web announcement there was a comarison of Z/VM and VMware running LINUX guests. I have the PDF presentation if you want it I can send it to you.. Or it is probably not too difficult to find on IBM's web site. -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] Behalf Of Alan Ackerman Sent: Thursday, October 30, 2008 6:25 PM To: IBMVM@LISTSERV.UARK.EDU Subject: I/O Overhead - z/VM versus VMWARE I got asked: Does z/VM impose non-insignificant overhead? Is it similar to VMware,= in which virtual I/O imposes significant overhead, but most processor and = memory access runs at close to native physical speed? I don't know anything about VMWARE so I could not answer the question. = I know that CCW Translation in VM costs significant cycles. I think FCP disks dedicated DASD fullpack minidisks small minidisks= . I would HOPE that the zSeries, with so much of virtualization built into = the hardware, would have lower costs than VMWARE, but I don't really kn= ow. Any takers? Are there any web sites that give performance comparisons VM versus VMWAR= E? Alan Ackerman= Alan (dot) Ackerman (at) Bank of America (dot) com
Re: Reliability of SFS?
On Fri, Oct 31, 2008 at 5:50 PM, Alan Altmark [EMAIL PROTECTED] wrote: I don't understand, Rob. Every time a program writes, it must check EVERY call it makes to ensure it worked. It can't just blindly continue. In fact, if the connection between client and server is broken, all changes to any open files are backed out, so the program must be restarted anyway. (Workunits, you know!) And if a minidisk is DETACHed and LINKed, CMS doesn't automatically ACCESS it for you. I don't know why you expect SFS to be different in this respect. When my ACCESS has completed with RC=0, I may assume the mini disk to be accessed correctly and remain there until I release it. It is making life very complicated when I must write my application to expect the disk to be pulled underneath my fingers any moment. I understand that one could also take a real disk away, but SFS servers tend to have more fingers in the pie. When my program code is on a remote file pool, I can't even trust that the disk with my EXEC is still there to read the remainder of the program code (including the error handler for example). This is not something you can expect me to handle. (I know I can execload the stuff, but it is making things complicated). -Rob
What's on a tape?
I've been asked to figure out what we have in our inventory of VM and VSE tapes, in preparation for moving to new drives/media. I'm expecting to write a PIPE that will do the required summary, more or less iterating to a final solution by operators will be able to use on our inventory of a few thousand tapes. Some of the expected formats include CMS and VSE labeled, unlabeled created mostly here, but with some that may have been created elsewhere. CMS formats include standard blocked unblocked, tape, vmfplc2, and various other CMS TAPE command local formats such as very old Perkin-Elmer compressed format and other 800-byte block file systems. In addition, I'll be recognizing specific formats such as SPSS system files, SPTAPE, etc. Does anyone have a starting point I might use in place of a bigger-than-I-want-project? ;-) cheers, wayne U Maine System (6 or 7 of 8 to 10 years into dismantling our VM system)
Re: What's on a tape?
Tapemap does an excellent job of Standard Label tapes and CMS VMFPLC2/TAPE dump datasets. Mount each tape, map it, rename the listing with the tape number. /Tom Kern Wayne T Smith wrote: I've been asked to figure out what we have in our inventory of VM and VSE tapes, in preparation for moving to new drives/media. I'm expecting to write a PIPE that will do the required summary, more or less iterating to a final solution by operators will be able to use on our inventory of a few thousand tapes. Some of the expected formats include CMS and VSE labeled, unlabeled created mostly here, but with some that may have been created elsewhere. CMS formats include standard blocked unblocked, tape, vmfplc2, and various other CMS TAPE command local formats such as very old Perkin-Elmer compressed format and other 800-byte block file systems. In addition, I'll be recognizing specific formats such as SPSS system files, SPTAPE, etc. Does anyone have a starting point I might use in place of a bigger-than-I-want-project? ;-) cheers, wayne U Maine System (6 or 7 of 8 to 10 years into dismantling our VM system)
Re: I/O Overhead - z/VM versus VMWARE
On Fri, 31 Oct 2008 08:28:45 -0500, Huegel, Thomas [EMAIL PROTECTED] wr ote: In the Z10-BC web announcement there was a comarison of Z/VM and VMware running LINUX guests. I have the PDF presentation if you want it I can send it to you.. Or it is probably not too difficult to find on IBM's web site. I could find nothing about VMWARE at IBM's site. Could you please send me a copy, or post a web link? Alan Ackerman Alan (dot) Ackerman (at) Bank of America (dot) com