Re: CPU time and zIIP
On the full speed box the delays will be real. On the faster kneecapped box the delays will be counted as part of the kneecapping. On Mon, Mar 11, 2019 at 3:40 PM Edward Finnell <000248cce9f3-dmarc-requ...@listserv.ua.edu> wrote: > > Is the overhead noticeable from a thruput standpoint? > > Say we have a 100 MSU box and a 200 MSU box capped at 100 will same workloads > complete in close proximity?In a message dated 3/4/2019 9:36:11 PM Central > Standard Time, edja...@phoenixsoftware.com writes: > So a better analogy might be that each CP is going around a race track > and pulls into the pits on each lap. After servicing, sub-capacity > models are held in the pits for awhile before they're allowed back on > the track. > > -- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- Mike A Schwab, Springfield IL USA Where do Forest Rangers go to get away from it all? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
Is the overhead noticeable from a thruput standpoint? Say we have a 100 MSU box and a 200 MSU box capped at 100 will same workloads complete in close proximity?In a message dated 3/4/2019 9:36:11 PM Central Standard Time, edja...@phoenixsoftware.com writes: So a better analogy might be that each CP is going around a race track and pulls into the pits on each lap. After servicing, sub-capacity models are held in the pits for awhile before they're allowed back on the track. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
thanks Rob, a PM from Patrick helped me find the error of my ways ! Carmen Vitullo - Original Message - From: "Rob Scott" To: IBM-MAIN@LISTSERV.UA.EDU Sent: Tuesday, March 5, 2019 11:09:54 AM Subject: Re: CPU time and zIIP Choose option 4 to show the fields and you go thru the list of columns using "enter". Once all columns have been shown, the next panel is the "Additional Information" menu that I mentioned. Also - some of the field names in the short field help are enabled for point-and-shoot for more information. Rob Scott -Original Message- From: IBM Mainframe Discussion List On Behalf Of Carmen Vitullo Sent: Tuesday, March 5, 2019 1:25 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: CPU time and zIIP I don't see the same HELP as you Rob, HELP or PF1 in DA panel gets me here Select a topic by number, or press Enter to view them in sequence. 1 - Introduction to the DA panel 2 - Syntax of the DA command 3 - Action characters: display output, cancel jobs, etc. 4 - Fields on the DA panel 5 - Overtyping fields to change their values 6 - Commands: limit jobs displayed, search, etc. These topics are displayed only if selected: 97 - What's new 98 - Search and navigate the help 99 - Messages for me, it's not a big deal, I can get the RMF/SMF data I need to get a good picture, somewhat historical, but I can get it. Carmen Vitullo - Original Message - From: "Rob Scott" To: IBM-MAIN@LISTSERV.UA.EDU Sent: Tuesday, March 5, 2019 4:39:14 AM Subject: Re: CPU time and zIIP Under the help section of the "DA" command, after the brief descriptions of the columns, there is an appendix section for detailed information on certain sections. It looks something like this : For more information, select a topic by number: 1 - Address space positions 2 - Swap out reason codes 3 - Server values 4 - Displaying alternate fields 5 - Changing the colors of fields 6 - Changing the order of fields 7 - CPU and SIO fields 8 - Scaling and abbreviations for values Choosing option 7, leads to several help panels describing in detail where SDSF sources the information and the calculations involved in showing the data, for example : CPU-Time and ECPU-Time columns: SDSF obtains the values for these columns from RMF, as follows: CPU-Time = ASCBEJST + ASCBSRBT + ASSBASST (source field R791TCPU) ECPU-Time = ASCBEJST + ASCBSRBT + ASSBPHTM (source field R791TCPC) where: ASCBEJST is elapsed job step time ASCBSRBT is accumulated SRB time ASSBASST is the CPU time consumed by preemptible class SRBs running on behalf of this address space, in milliseconds ASSBPHTM is the CPU time consumed by preemptible class SRBs running in this address space, in milliseconds (threads plus enclaves) GCP-Time, zAAP-Time and zACP-Time are not normalized. SDSF obtains the values for these columns from RMF: GCP-Time source field is R791TCPU zAAP-Time source field is R791TIFA zACP-Time source field is R791TIFC zAAP-NTime is normalized to the slower CP, to facilitate comparing values. The normalization uses fields from RMF, as follows: R791TIFA x R791NFFI --- 256 ..etc etc Rob Scott Rocket Software -Original Message- From: IBM Mainframe Discussion List On Behalf Of Seymour J Metz Sent: Monday, March 4, 2019 5:10 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: CPU time and zIIP The last time that I looked SDSF did not document the actual meanings of all of the fields that it displayed. A lot of these discussions could go away if IBM spelled out the in SDSF documentation the actual meaning of each "CPU" field instead of using ambiguous boilerplate language. -- Shmuel (Seymour J.) Metz https://nam01.safelinks.protection.outlook.com/?url=http:%2F%2Fmason.gmu.edu%2F~smetz3data=02%7C01%7CRScott%40ROCKETSOFTWARE.COM%7C9742b37cf34f48ed751108d6a16df68f%7C79544c1eed224879a082b67a9a672aae%7C0%7C0%7C636873891015867170sdata=Ai0Sh6AHgWFIhfyfNAR7m4gPqPlfntm4pua%2BbO2om%2Fs%3Dreserved=0 From: IBM Mainframe Discussion List on behalf of Carmen Vitullo Sent: Monday, March 4, 2019 9:10 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: CPU time and zIIP bad assumption on my part that SDSF show's CPU and zIIP percent for each address space zAAP-Time zACP-Time GCP-Use% zAAP-Use% SzAAP% SzIIP% P Carmen Vitullo - Original Message - From: "Carmen Vitullo" To: IBM-MAIN@LISTSERV.UA.EDU Sent: Friday, March 1, 2019 7:28:54 AM Subject: Re: CPU time and zIIP Carmen Vitullo - Original Message - From: "Peter Relson" To: IBM-MAIN@LISTSERV.UA.EDU Sent: Friday, March 1, 2019 7:15:51 AM Subject: Re: CPU time and zIIP I'm obviously still not understanding what you think is amiss. adding up the Gcpu time and Ziip time, to show all CP TIME but seeing that I mi
Re: CPU time and zIIP
Choose option 4 to show the fields and you go thru the list of columns using "enter". Once all columns have been shown, the next panel is the "Additional Information" menu that I mentioned. Also - some of the field names in the short field help are enabled for point-and-shoot for more information. Rob Scott -Original Message- From: IBM Mainframe Discussion List On Behalf Of Carmen Vitullo Sent: Tuesday, March 5, 2019 1:25 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: CPU time and zIIP I don't see the same HELP as you Rob, HELP or PF1 in DA panel gets me here Select a topic by number, or press Enter to view them in sequence. 1 - Introduction to the DA panel 2 - Syntax of the DA command 3 - Action characters: display output, cancel jobs, etc. 4 - Fields on the DA panel 5 - Overtyping fields to change their values 6 - Commands: limit jobs displayed, search, etc. These topics are displayed only if selected: 97 - What's new 98 - Search and navigate the help 99 - Messages for me, it's not a big deal, I can get the RMF/SMF data I need to get a good picture, somewhat historical, but I can get it. Carmen Vitullo - Original Message - From: "Rob Scott" To: IBM-MAIN@LISTSERV.UA.EDU Sent: Tuesday, March 5, 2019 4:39:14 AM Subject: Re: CPU time and zIIP Under the help section of the "DA" command, after the brief descriptions of the columns, there is an appendix section for detailed information on certain sections. It looks something like this : For more information, select a topic by number: 1 - Address space positions 2 - Swap out reason codes 3 - Server values 4 - Displaying alternate fields 5 - Changing the colors of fields 6 - Changing the order of fields 7 - CPU and SIO fields 8 - Scaling and abbreviations for values Choosing option 7, leads to several help panels describing in detail where SDSF sources the information and the calculations involved in showing the data, for example : CPU-Time and ECPU-Time columns: SDSF obtains the values for these columns from RMF, as follows: CPU-Time = ASCBEJST + ASCBSRBT + ASSBASST (source field R791TCPU) ECPU-Time = ASCBEJST + ASCBSRBT + ASSBPHTM (source field R791TCPC) where: ASCBEJST is elapsed job step time ASCBSRBT is accumulated SRB time ASSBASST is the CPU time consumed by preemptible class SRBs running on behalf of this address space, in milliseconds ASSBPHTM is the CPU time consumed by preemptible class SRBs running in this address space, in milliseconds (threads plus enclaves) GCP-Time, zAAP-Time and zACP-Time are not normalized. SDSF obtains the values for these columns from RMF: GCP-Time source field is R791TCPU zAAP-Time source field is R791TIFA zACP-Time source field is R791TIFC zAAP-NTime is normalized to the slower CP, to facilitate comparing values. The normalization uses fields from RMF, as follows: R791TIFA x R791NFFI --- 256 ..etc etc Rob Scott Rocket Software -Original Message- From: IBM Mainframe Discussion List On Behalf Of Seymour J Metz Sent: Monday, March 4, 2019 5:10 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: CPU time and zIIP The last time that I looked SDSF did not document the actual meanings of all of the fields that it displayed. A lot of these discussions could go away if IBM spelled out the in SDSF documentation the actual meaning of each "CPU" field instead of using ambiguous boilerplate language. -- Shmuel (Seymour J.) Metz https://nam01.safelinks.protection.outlook.com/?url=http:%2F%2Fmason.gmu.edu%2F~smetz3data=02%7C01%7CRScott%40ROCKETSOFTWARE.COM%7C9742b37cf34f48ed751108d6a16df68f%7C79544c1eed224879a082b67a9a672aae%7C0%7C0%7C636873891015867170sdata=Ai0Sh6AHgWFIhfyfNAR7m4gPqPlfntm4pua%2BbO2om%2Fs%3Dreserved=0 From: IBM Mainframe Discussion List on behalf of Carmen Vitullo Sent: Monday, March 4, 2019 9:10 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: CPU time and zIIP bad assumption on my part that SDSF show's CPU and zIIP percent for each address space zAAP-Time zACP-Time GCP-Use% zAAP-Use% SzAAP% SzIIP% P Carmen Vitullo - Original Message - From: "Carmen Vitullo" To: IBM-MAIN@LISTSERV.UA.EDU Sent: Friday, March 1, 2019 7:28:54 AM Subject: Re: CPU time and zIIP Carmen Vitullo - Original Message - From: "Peter Relson" To: IBM-MAIN@LISTSERV.UA.EDU Sent: Friday, March 1, 2019 7:15:51 AM Subject: Re: CPU time and zIIP I'm obviously still not understanding what you think is amiss. adding up the Gcpu time and Ziip time, to show all CP TIME but seeing that I misread, misunderstood the doc and your previous responses and from the statement belowas Gilda Radner would saynevermind :( for me I'd like to see SDSF's CPUtime to include all time, GCPU+IIP We have already said that it does. Peter Relson z/OS Core Technology Design is that true for CPU percent also? thanks for provide
Re: CPU time and zIIP
I don't see the same HELP as you Rob, HELP or PF1 in DA panel gets me here Select a topic by number, or press Enter to view them in sequence. 1 - Introduction to the DA panel 2 - Syntax of the DA command 3 - Action characters: display output, cancel jobs, etc. 4 - Fields on the DA panel 5 - Overtyping fields to change their values 6 - Commands: limit jobs displayed, search, etc. These topics are displayed only if selected: 97 - What's new 98 - Search and navigate the help 99 - Messages for me, it's not a big deal, I can get the RMF/SMF data I need to get a good picture, somewhat historical, but I can get it. Carmen Vitullo - Original Message - From: "Rob Scott" To: IBM-MAIN@LISTSERV.UA.EDU Sent: Tuesday, March 5, 2019 4:39:14 AM Subject: Re: CPU time and zIIP Under the help section of the "DA" command, after the brief descriptions of the columns, there is an appendix section for detailed information on certain sections. It looks something like this : For more information, select a topic by number: 1 - Address space positions 2 - Swap out reason codes 3 - Server values 4 - Displaying alternate fields 5 - Changing the colors of fields 6 - Changing the order of fields 7 - CPU and SIO fields 8 - Scaling and abbreviations for values Choosing option 7, leads to several help panels describing in detail where SDSF sources the information and the calculations involved in showing the data, for example : CPU-Time and ECPU-Time columns: SDSF obtains the values for these columns from RMF, as follows: CPU-Time = ASCBEJST + ASCBSRBT + ASSBASST (source field R791TCPU) ECPU-Time = ASCBEJST + ASCBSRBT + ASSBPHTM (source field R791TCPC) where: ASCBEJST is elapsed job step time ASCBSRBT is accumulated SRB time ASSBASST is the CPU time consumed by preemptible class SRBs running on behalf of this address space, in milliseconds ASSBPHTM is the CPU time consumed by preemptible class SRBs running in this address space, in milliseconds (threads plus enclaves) GCP-Time, zAAP-Time and zACP-Time are not normalized. SDSF obtains the values for these columns from RMF: GCP-Time source field is R791TCPU zAAP-Time source field is R791TIFA zACP-Time source field is R791TIFC zAAP-NTime is normalized to the slower CP, to facilitate comparing values. The normalization uses fields from RMF, as follows: R791TIFA x R791NFFI --- 256 ..etc etc Rob Scott Rocket Software -Original Message- From: IBM Mainframe Discussion List On Behalf Of Seymour J Metz Sent: Monday, March 4, 2019 5:10 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: CPU time and zIIP The last time that I looked SDSF did not document the actual meanings of all of the fields that it displayed. A lot of these discussions could go away if IBM spelled out the in SDSF documentation the actual meaning of each "CPU" field instead of using ambiguous boilerplate language. -- Shmuel (Seymour J.) Metz https://nam01.safelinks.protection.outlook.com/?url=http:%2F%2Fmason.gmu.edu%2F~smetz3data=02%7C01%7CRScott%40ROCKETSOFTWARE.COM%7Ca18442b199c54a67875908d6a0ea1a7a%7C79544c1eed224879a082b67a9a672aae%7C0%7C0%7C636873324684797715sdata=VlY%2BTXoUkEyV15PpoJZQy0Na%2Ft2In%2BXQmpsNTkMe1Jo%3Dreserved=0 From: IBM Mainframe Discussion List on behalf of Carmen Vitullo Sent: Monday, March 4, 2019 9:10 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: CPU time and zIIP bad assumption on my part that SDSF show's CPU and zIIP percent for each address space zAAP-Time zACP-Time GCP-Use% zAAP-Use% SzAAP% SzIIP% P Carmen Vitullo - Original Message - From: "Carmen Vitullo" To: IBM-MAIN@LISTSERV.UA.EDU Sent: Friday, March 1, 2019 7:28:54 AM Subject: Re: CPU time and zIIP Carmen Vitullo - Original Message - From: "Peter Relson" To: IBM-MAIN@LISTSERV.UA.EDU Sent: Friday, March 1, 2019 7:15:51 AM Subject: Re: CPU time and zIIP I'm obviously still not understanding what you think is amiss. adding up the Gcpu time and Ziip time, to show all CP TIME but seeing that I misread, misunderstood the doc and your previous responses and from the statement belowas Gilda Radner would saynevermind :( for me I'd like to see SDSF's CPUtime to include all time, GCPU+IIP We have already said that it does. Peter Relson z/OS Core Technology Design is that true for CPU percent also? thanks for provided my the clue by four I needed to get this into my thick skull Carmen -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists.
Re: CPU time and zIIP
Under the help section of the "DA" command, after the brief descriptions of the columns, there is an appendix section for detailed information on certain sections. It looks something like this : For more information, select a topic by number: 1 - Address space positions 2 - Swap out reason codes 3 - Server values 4 - Displaying alternate fields 5 - Changing the colors of fields 6 - Changing the order of fields 7 - CPU and SIO fields 8 - Scaling and abbreviations for values Choosing option 7, leads to several help panels describing in detail where SDSF sources the information and the calculations involved in showing the data, for example : CPU-Time and ECPU-Time columns: SDSF obtains the values for these columns from RMF, as follows: CPU-Time = ASCBEJST + ASCBSRBT + ASSBASST (source field R791TCPU) ECPU-Time = ASCBEJST + ASCBSRBT + ASSBPHTM (source field R791TCPC) where: ASCBEJST is elapsed job step time ASCBSRBT is accumulated SRB time ASSBASST is the CPU time consumed by preemptible class SRBs running on behalf of this address space, in milliseconds ASSBPHTM is the CPU time consumed by preemptible class SRBs running in this address space, in milliseconds (threads plus enclaves) GCP-Time, zAAP-Time and zACP-Time are not normalized. SDSF obtains the values for these columns from RMF: GCP-Time source field is R791TCPU zAAP-Time source field is R791TIFA zACP-Time source field is R791TIFC zAAP-NTime is normalized to the slower CP, to facilitate comparing values. The normalization uses fields from RMF, as follows: R791TIFA x R791NFFI --- 256 ..etc etc Rob Scott Rocket Software -Original Message- From: IBM Mainframe Discussion List On Behalf Of Seymour J Metz Sent: Monday, March 4, 2019 5:10 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: CPU time and zIIP The last time that I looked SDSF did not document the actual meanings of all of the fields that it displayed. A lot of these discussions could go away if IBM spelled out the in SDSF documentation the actual meaning of each "CPU" field instead of using ambiguous boilerplate language. -- Shmuel (Seymour J.) Metz https://nam01.safelinks.protection.outlook.com/?url=http:%2F%2Fmason.gmu.edu%2F~smetz3data=02%7C01%7CRScott%40ROCKETSOFTWARE.COM%7Ca18442b199c54a67875908d6a0ea1a7a%7C79544c1eed224879a082b67a9a672aae%7C0%7C0%7C636873324684797715sdata=VlY%2BTXoUkEyV15PpoJZQy0Na%2Ft2In%2BXQmpsNTkMe1Jo%3Dreserved=0 From: IBM Mainframe Discussion List on behalf of Carmen Vitullo Sent: Monday, March 4, 2019 9:10 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: CPU time and zIIP bad assumption on my part that SDSF show's CPU and zIIP percent for each address space zAAP-Time zACP-Time GCP-Use% zAAP-Use% SzAAP% SzIIP% P Carmen Vitullo - Original Message - From: "Carmen Vitullo" To: IBM-MAIN@LISTSERV.UA.EDU Sent: Friday, March 1, 2019 7:28:54 AM Subject: Re: CPU time and zIIP Carmen Vitullo - Original Message - From: "Peter Relson" To: IBM-MAIN@LISTSERV.UA.EDU Sent: Friday, March 1, 2019 7:15:51 AM Subject: Re: CPU time and zIIP I'm obviously still not understanding what you think is amiss. adding up the Gcpu time and Ziip time, to show all CP TIME but seeing that I misread, misunderstood the doc and your previous responses and from the statement belowas Gilda Radner would saynevermind :( for me I'd like to see SDSF's CPUtime to include all time, GCPU+IIP We have already said that it does. Peter Relson z/OS Core Technology Design is that true for CPU percent also? thanks for provided my the clue by four I needed to get this into my thick skull Carmen -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN Rocket Software, Inc. and subsidiaries ■ 77 Fourth Avenue, Waltham MA 02451 ■ Main Office Toll Free Number: +1 855.577.4323 Contact Customer Support: https://my.rocketsoftware.com/RocketCommunity/RCEmailSupport Unsubscribe from Marketing Messages/Manage Your Subscription Preferences
Re: CPU time and zIIP
The last time that I looked SDSF did not document the actual meanings of all of the fields that it displayed. A lot of these discussions could go away if IBM spelled out the in SDSF documentation the actual meaning of each "CPU" field instead of using ambiguous boilerplate language. -- Shmuel (Seymour J.) Metz http://mason.gmu.edu/~smetz3 From: IBM Mainframe Discussion List on behalf of Carmen Vitullo Sent: Monday, March 4, 2019 9:10 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: CPU time and zIIP bad assumption on my part that SDSF show's CPU and zIIP percent for each address space zAAP-Time zACP-Time GCP-Use% zAAP-Use% SzAAP% SzIIP% P Carmen Vitullo - Original Message - From: "Carmen Vitullo" To: IBM-MAIN@LISTSERV.UA.EDU Sent: Friday, March 1, 2019 7:28:54 AM Subject: Re: CPU time and zIIP Carmen Vitullo - Original Message - From: "Peter Relson" To: IBM-MAIN@LISTSERV.UA.EDU Sent: Friday, March 1, 2019 7:15:51 AM Subject: Re: CPU time and zIIP I'm obviously still not understanding what you think is amiss. adding up the Gcpu time and Ziip time, to show all CP TIME but seeing that I misread, misunderstood the doc and your previous responses and from the statement belowas Gilda Radner would saynevermind :( for me I'd like to see SDSF's CPUtime to include all time, GCPU+IIP We have already said that it does. Peter Relson z/OS Core Technology Design is that true for CPU percent also? thanks for provided my the clue by four I needed to get this into my thick skull Carmen -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
On 3/4/2019 5:39 AM, R.S. wrote: W dniu 2019-03-03 o 17:07, Christopher Y. Blaicher pisze: ZIIP and ZAAP processors always run at full speed, even when running on a sub-capacity box. One thing, among many, I don't know is how IBM implements sub-capacity. Slow the clock speed? Skip cycles? All processors, including subcapacity CPs run at the same speed, which is different for big box and small (BC) box. How subcapacity CP makes the subcapacity? By stealing cycles, imagine there's hidden LPAR with some fixed weigth. IMHO that's not a good analogy. The cycles aren't "stolen" in a manner at all similar to a so-called "soak" LPAR. Each CP has a timed "service loop" with a housekeeping routine -- implemented in millicode -- that handles interrupts, SIGP instructions, etc. When the housekeeping is finished, the CP runs instructions for the remainder of the interval. One of the last things in the housekeeping routine is a coded loop (probably using an ordinary BCT instruction) that iterates "N" times where the value of "N" is a constant that varies depending on the sub-capacity model. So a better analogy might be that each CP is going around a race track and pulls into the pits on each lap. After servicing, sub-capacity models are held in the pits for awhile before they're allowed back on the track. -- Phoenix Software International Edward E. Jaffe 831 Parkview Drive North El Segundo, CA 90245 https://www.phoenixsoftware.com/ This e-mail message, including any attachments, appended messages and the information contained therein, is for the sole use of the intended recipient(s). If you are not an intended recipient or have otherwise received this email message in error, any use, dissemination, distribution, review, storage or copying of this e-mail message and the information contained therein is strictly prohibited. If you are not an intended recipient, please contact the sender by reply e-mail and destroy all copies of this email message and do not otherwise utilize or retain this email message or any or all of the information contained therein. Although this email message and any attachments or appended messages are believed to be free of any virus or other defect that might affect any computer system into which it is received and opened, it is the responsibility of the recipient to ensure that it is virus free and no responsibility is accepted by the sender for any loss or damage arising in any way from its opening or use. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
And that is precisely why I asked the question: To get people thinking about how this might show up instrumentation. (And no I didn't think there was a dummy LPAR.) Cheers, Martin Martin Packer zChampion, Systems Investigator & Performance Troubleshooter, IBM +44-7802-245-584 email: martin_pac...@uk.ibm.com Twitter / Facebook IDs: MartinPacker Blog: https://www.ibm.com/developerworks/mydeveloperworks/blogs/MartinPacker Podcast Series (With Marna Walle): https://developer.ibm.com/tv/mpt/or https://itunes.apple.com/gb/podcast/mainframe-performance-topics/id1127943573?mt=2 Youtube channel: https://www.youtube.com/channel/UCu_65HaYgksbF6Q8SQ4oOvA From: Jim Mulder To: IBM-MAIN@LISTSERV.UA.EDU Date: 04/03/2019 15:39 Subject:Re: CPU time and zIIP Sent by:IBM Mainframe Discussion List There is no "dummy LPAR". Millicode periodically executes a loop to waste some time. The logical processor remains dispatched to the physical processor while that is happening, so the wasted time is included in the CPU Timer for the logical processor, and thus is charged to the dispatched work unit, as it should be. Jim Mulder z/OS Diagnosis, Design, Development, Test IBM Corp. Poughkeepsie NY > It is simple: clock remains the same for any subcapacity. Simply some > cycles are taken by "dummy LPAR". > How does it relate to CPI? It does not affect... -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
Dummy LPAR or milicode waste the cycles - nevermind. My point was there are some cycles lost, not longer (slower) cycles. Regards -- Radoslaw Skorupka Lodz, Poland W dniu 2019-03-04 o 16:39, Jim Mulder pisze: There is no "dummy LPAR". Millicode periodically executes a loop to waste some time. The logical processor remains dispatched to the physical processor while that is happening, so the wasted time is included in the CPU Timer for the logical processor, and thus is charged to the dispatched work unit, as it should be. Jim Mulder z/OS Diagnosis, Design, Development, Test IBM Corp. Poughkeepsie NY It is simple: clock remains the same for any subcapacity. Simply some cycles are taken by "dummy LPAR". How does it relate to CPI? It does not affect... -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN . == Jeśli nie jesteś adresatem tej wiadomości: - powiadom nas o tym w mailu zwrotnym (dziękujemy!), - usuń trwale tę wiadomość (i wszystkie kopie, które wydrukowałeś lub zapisałeś na dysku). Wiadomość ta może zawierać chronione prawem informacje, które może wykorzystać tylko adresat.Przypominamy, że każdy, kto rozpowszechnia (kopiuje, rozprowadza) tę wiadomość lub podejmuje podobne działania, narusza prawo i może podlegać karze. mBank S.A. z siedzibą w Warszawie, ul. Senatorska 18, 00-950 Warszawa,www.mBank.pl, e-mail: kont...@mbank.pl. Sąd Rejonowy dla m. st. Warszawy XII Wydział Gospodarczy Krajowego Rejestru Sądowego, KRS 025237, NIP: 526-021-50-88. Kapitał zakładowy (opłacony w całości) według stanu na 01.01.2018 r. wynosi 169.248.488 złotych. If you are not the addressee of this message: - let us know by replying to this e-mail (thank you!), - delete this message permanently (including all the copies which you have printed out or saved). This message may contain legally protected information, which may be used exclusively by the addressee.Please be reminded that anyone who disseminates (copies, distributes) this message or takes any similar action, violates the law and may be penalised. mBank S.A. with its registered office in Warsaw, ul. Senatorska 18, 00-950 Warszawa,www.mBank.pl, e-mail: kont...@mbank.pl. District Court for the Capital City of Warsaw, 12th Commercial Division of the National Court Register, KRS 025237, NIP: 526-021-50-88. Fully paid-up share capital amounting to PLN 169,248,488 as at 1 January 2018. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
There is no "dummy LPAR". Millicode periodically executes a loop to waste some time. The logical processor remains dispatched to the physical processor while that is happening, so the wasted time is included in the CPU Timer for the logical processor, and thus is charged to the dispatched work unit, as it should be. Jim Mulder z/OS Diagnosis, Design, Development, Test IBM Corp. Poughkeepsie NY > It is simple: clock remains the same for any subcapacity. Simply some > cycles are taken by "dummy LPAR". > How does it relate to CPI? It does not affect... -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
Why are you asking? It is simple: clock remains the same for any subcapacity. Simply some cycles are taken by "dummy LPAR". How does it relate to CPI? It does not affect... -- Radoslaw Skorupka Lodz, Poland W dniu 2019-03-04 o 14:51, Martin Packer pisze: Would you expect Cycles-Per-Instruction (CPI) in SMF 113 to be larger for a subcapacity machine? Cheers, Martin Martin Packer zChampion, Systems Investigator & Performance Troubleshooter, IBM +44-7802-245-584 email: martin_pac...@uk.ibm.com Twitter / Facebook IDs: MartinPacker Blog: https://www.ibm.com/developerworks/mydeveloperworks/blogs/MartinPacker Podcast Series (With Marna Walle): https://developer.ibm.com/tv/mpt/or https://itunes.apple.com/gb/podcast/mainframe-performance-topics/id1127943573?mt=2 Youtube channel: https://www.youtube.com/channel/UCu_65HaYgksbF6Q8SQ4oOvA From: "R.S." To: IBM-MAIN@LISTSERV.UA.EDU Date: 04/03/2019 13:39 Subject: Re: CPU time and zIIP Sent by:IBM Mainframe Discussion List W dniu 2019-03-03 o 17:07, Christopher Y. Blaicher pisze: ZIIP and ZAAP processors always run at full speed, even when running on a sub-capacity box. One thing, among many, I don't know is how IBM implements sub-capacity. Slow the clock speed? Skip cycles? All processors, including subcapacity CPs run at the same speed, which is different for big box and small (BC) box. How subcapacity CP makes the subcapacity? By stealing cycles, imagine there's hidden LPAR with some fixed weigth. Note, the processors are actually cores in PU, AFAIK there is no way the cores would work at different speeds within the PU. Note 2, the same method (stealing cycles, not slowing down the clock) was in use for RAn machines in 9672 era. == Jeśli nie jesteś adresatem tej wiadomości: - powiadom nas o tym w mailu zwrotnym (dziękujemy!), - usuń trwale tę wiadomość (i wszystkie kopie, które wydrukowałeś lub zapisałeś na dysku). Wiadomość ta może zawierać chronione prawem informacje, które może wykorzystać tylko adresat.Przypominamy, że każdy, kto rozpowszechnia (kopiuje, rozprowadza) tę wiadomość lub podejmuje podobne działania, narusza prawo i może podlegać karze. mBank S.A. z siedzibą w Warszawie, ul. Senatorska 18, 00-950 Warszawa,www.mBank.pl, e-mail: kont...@mbank.pl. Sąd Rejonowy dla m. st. Warszawy XII Wydział Gospodarczy Krajowego Rejestru Sądowego, KRS 025237, NIP: 526-021-50-88. Kapitał zakładowy (opłacony w całości) według stanu na 01.01.2018 r. wynosi 169.248.488 złotych. If you are not the addressee of this message: - let us know by replying to this e-mail (thank you!), - delete this message permanently (including all the copies which you have printed out or saved). This message may contain legally protected information, which may be used exclusively by the addressee.Please be reminded that anyone who disseminates (copies, distributes) this message or takes any similar action, violates the law and may be penalised. mBank S.A. with its registered office in Warsaw, ul. Senatorska 18, 00-950 Warszawa,www.mBank.pl, e-mail: kont...@mbank.pl. District Court for the Capital City of Warsaw, 12th Commercial Division of the National Court Register, KRS 025237, NIP: 526-021-50-88. Fully paid-up share capital amounting to PLN 169,248,488 as at 1 January 2018. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
bad assumption on my part that SDSF show's CPU and zIIP percent for each address space zAAP-Time zACP-Time GCP-Use% zAAP-Use% SzAAP% SzIIP% P Carmen Vitullo - Original Message - From: "Carmen Vitullo" To: IBM-MAIN@LISTSERV.UA.EDU Sent: Friday, March 1, 2019 7:28:54 AM Subject: Re: CPU time and zIIP Carmen Vitullo - Original Message - From: "Peter Relson" To: IBM-MAIN@LISTSERV.UA.EDU Sent: Friday, March 1, 2019 7:15:51 AM Subject: Re: CPU time and zIIP I'm obviously still not understanding what you think is amiss. adding up the Gcpu time and Ziip time, to show all CP TIME but seeing that I misread, misunderstood the doc and your previous responses and from the statement belowas Gilda Radner would saynevermind :( for me I'd like to see SDSF's CPUtime to include all time, GCPU+IIP We have already said that it does. Peter Relson z/OS Core Technology Design is that true for CPU percent also? thanks for provided my the clue by four I needed to get this into my thick skull Carmen -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
Would you expect Cycles-Per-Instruction (CPI) in SMF 113 to be larger for a subcapacity machine? Cheers, Martin Martin Packer zChampion, Systems Investigator & Performance Troubleshooter, IBM +44-7802-245-584 email: martin_pac...@uk.ibm.com Twitter / Facebook IDs: MartinPacker Blog: https://www.ibm.com/developerworks/mydeveloperworks/blogs/MartinPacker Podcast Series (With Marna Walle): https://developer.ibm.com/tv/mpt/or https://itunes.apple.com/gb/podcast/mainframe-performance-topics/id1127943573?mt=2 Youtube channel: https://www.youtube.com/channel/UCu_65HaYgksbF6Q8SQ4oOvA From: "R.S." To: IBM-MAIN@LISTSERV.UA.EDU Date: 04/03/2019 13:39 Subject: Re: CPU time and zIIP Sent by:IBM Mainframe Discussion List W dniu 2019-03-03 o 17:07, Christopher Y. Blaicher pisze: > ZIIP and ZAAP processors always run at full speed, even when running on a sub-capacity box. > One thing, among many, I don't know is how IBM implements sub-capacity. Slow the clock speed? Skip cycles? All processors, including subcapacity CPs run at the same speed, which is different for big box and small (BC) box. How subcapacity CP makes the subcapacity? By stealing cycles, imagine there's hidden LPAR with some fixed weigth. Note, the processors are actually cores in PU, AFAIK there is no way the cores would work at different speeds within the PU. Note 2, the same method (stealing cycles, not slowing down the clock) was in use for RAn machines in 9672 era. -- Radoslaw Skorupka Lodz, Poland == Jeśli nie jesteś adresatem tej wiadomości: - powiadom nas o tym w mailu zwrotnym (dziękujemy!), - usuń trwale tę wiadomość (i wszystkie kopie, które wydrukowałeś lub zapisałeś na dysku). Wiadomość ta może zawierać chronione prawem informacje, które może wykorzystać tylko adresat.Przypominamy, że każdy, kto rozpowszechnia (kopiuje, rozprowadza) tę wiadomość lub podejmuje podobne działania, narusza prawo i może podlegać karze. mBank S.A. z siedzibą w Warszawie, ul. Senatorska 18, 00-950 Warszawa, www.mBank.pl, e-mail: kont...@mbank.pl. Sąd Rejonowy dla m. st. Warszawy XII Wydział Gospodarczy Krajowego Rejestru Sądowego, KRS 025237, NIP: 526-021-50-88. Kapitał zakładowy (opłacony w całości) według stanu na 01.01.2018 r. wynosi 169.248.488 złotych. If you are not the addressee of this message: - let us know by replying to this e-mail (thank you!), - delete this message permanently (including all the copies which you have printed out or saved). This message may contain legally protected information, which may be used exclusively by the addressee.Please be reminded that anyone who disseminates (copies, distributes) this message or takes any similar action, violates the law and may be penalised. mBank S.A. with its registered office in Warsaw, ul. Senatorska 18, 00-950 Warszawa,www.mBank.pl, e-mail: kont...@mbank.pl. District Court for the Capital City of Warsaw, 12th Commercial Division of the National Court Register, KRS 025237, NIP: 526-021-50-88. Fully paid-up share capital amounting to PLN 169,248,488 as at 1 January 2018. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
W dniu 2019-03-03 o 17:07, Christopher Y. Blaicher pisze: ZIIP and ZAAP processors always run at full speed, even when running on a sub-capacity box. One thing, among many, I don't know is how IBM implements sub-capacity. Slow the clock speed? Skip cycles? All processors, including subcapacity CPs run at the same speed, which is different for big box and small (BC) box. How subcapacity CP makes the subcapacity? By stealing cycles, imagine there's hidden LPAR with some fixed weigth. Note, the processors are actually cores in PU, AFAIK there is no way the cores would work at different speeds within the PU. Note 2, the same method (stealing cycles, not slowing down the clock) was in use for RAn machines in 9672 era. -- Radoslaw Skorupka Lodz, Poland == Jeśli nie jesteś adresatem tej wiadomości: - powiadom nas o tym w mailu zwrotnym (dziękujemy!), - usuń trwale tę wiadomość (i wszystkie kopie, które wydrukowałeś lub zapisałeś na dysku). Wiadomość ta może zawierać chronione prawem informacje, które może wykorzystać tylko adresat.Przypominamy, że każdy, kto rozpowszechnia (kopiuje, rozprowadza) tę wiadomość lub podejmuje podobne działania, narusza prawo i może podlegać karze. mBank S.A. z siedzibą w Warszawie, ul. Senatorska 18, 00-950 Warszawa,www.mBank.pl, e-mail: kont...@mbank.pl. Sąd Rejonowy dla m. st. Warszawy XII Wydział Gospodarczy Krajowego Rejestru Sądowego, KRS 025237, NIP: 526-021-50-88. Kapitał zakładowy (opłacony w całości) według stanu na 01.01.2018 r. wynosi 169.248.488 złotych. If you are not the addressee of this message: - let us know by replying to this e-mail (thank you!), - delete this message permanently (including all the copies which you have printed out or saved). This message may contain legally protected information, which may be used exclusively by the addressee.Please be reminded that anyone who disseminates (copies, distributes) this message or takes any similar action, violates the law and may be penalised. mBank S.A. with its registered office in Warsaw, ul. Senatorska 18, 00-950 Warszawa,www.mBank.pl, e-mail: kont...@mbank.pl. District Court for the Capital City of Warsaw, 12th Commercial Division of the National Court Register, KRS 025237, NIP: 526-021-50-88. Fully paid-up share capital amounting to PLN 169,248,488 as at 1 January 2018. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
Not in RMF, in SDSF. You are correct the OP question was about cput time, and the answer was, cputime both gcpu and ziip cpu were combined on the SDSF DA panel, CPUTIME field, SDSF also shows cpu percent, so just a curiosity if SDSF combines the CPU percent, usage for each address space combined also, just curious. I understand the differences in the engines and the speed they run as compared to a GPCPU, so maybe apples and oranges cannot be combined, if that's the reason why they are not. Carmen Vitullo - Original Message - From: "Peter Relson" To: IBM-MAIN@LISTSERV.UA.EDU Sent: Saturday, March 2, 2019 10:20:36 AM Subject: Re: CPU time and zIIP >is that true for CPU percent also? The original post and my answers were about SDSF. They did not show (to the best of my eyesight) or discuss any percentage fields. Are you now asking a question about a percentage shown in RMF? I don't pretend to know anything about what RMF displays, but it would likely not be a good idea to try to lump together busy percentages pertaining to standard CPs and zIIPs because the two types of processors are intended to have different characteristics in terms of how busy it is appropriate for them to be. Peter Relson z/OS Core Technology Design -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
One regular misconception about (cycle time), irrespective of the type of processor the 'speed/cycle time' of ALL the processors is the SAME I'd disagree a bit. I think that the "misconception" is of conflating "speed" with "cycle time". It is true that the machine cycle time for all of the processor types is the same. But the speed might not be, depending on configuration. So it is best to avoid using a term such as "speed/cycle time". Peter Relson z/OS Core Technology Design -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
Back in June 2018 there was discussion on this list that indicated that the forward progress on a subcapacity processor is suspended sufficiently often for a small time interval (4 µsec?) to degrade the average SU capacity of the processor to the licensed value. It would seem logical to assume such suspension would have to occur on clock cycle boundaries and be for one or more clock cycles. That would mean that while the processor is making progress, it would be running at full clock speed, but there would be some clock cycles where no progress occurred. From the users perspective and program code execution speed perspective it would appear as if the effective clock speed were reduced even though the physical clock speed was unchanged. This was the same design concept used to produce a sub-capacity IBM 407 Accounting machine back in the 1950's: the IBM 407 ran at 150 cards/minute, but a degraded and presumably cheaper-to-lease model was available that paused every third machine cycle, reducing the effective speed to 100 cards/minute. Joel C. Ewing On 3/3/19 10:07 AM, Christopher Y. Blaicher wrote: > ZIIP and ZAAP processors always run at full speed, even when running on a > sub-capacity box. > One thing, among many, I don't know is how IBM implements sub-capacity. Slow > the clock speed? Skip cycles? > > Chris Blaicher > Technical Architect > Syncsort, Inc. > > > -Original Message- > From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On > Behalf Of Parwez > Sent: Sunday, March 3, 2019 6:05 AM > To: IBM-MAIN@LISTSERV.UA.EDU > Subject: Re: CPU time and zIIP > > One regular misconception about (cycle time), irrespective of the type of > processor the 'speed/cycle time' of ALL the processors is the SAME. CPs with > CAPACITY setting of 7xx (Zxx in case of the 'BC' class system) are FULL > capacity. Others are sub-capacity. So if the workload system hasn't enough > capacity, then it might not run as well as one with abundance. So from a > cycle point (speed?), zIIPs are not faster. > > Parwez Hamid > > ... -- Joel C. Ewing -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
Just to make sure I myself haven't caused any confusion, as previously stated, irrespective of the type of processor i.e SAP, CP, ICF, IFL, zIIP (and systems which had zAAP) the 'speed/cycle time' of ALL the different processor types is always the SAME (this includes sub-capacity CPs). Only CPs have the sub-capacity settings. Parwez Hamid From: IBM Mainframe Discussion List on behalf of Christopher Y. Blaicher Sent: 03 March 2019 16:07 To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: CPU time and zIIP ZIIP and ZAAP processors always run at full speed, even when running on a sub-capacity box. One thing, among many, I don't know is how IBM implements sub-capacity. Slow the clock speed? Skip cycles? Chris Blaicher Technical Architect Syncsort, Inc. -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Parwez Sent: Sunday, March 3, 2019 6:05 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: CPU time and zIIP One regular misconception about (cycle time), irrespective of the type of processor the 'speed/cycle time' of ALL the processors is the SAME. CPs with CAPACITY setting of 7xx (Zxx in case of the 'BC' class system) are FULL capacity. Others are sub-capacity. So if the workload system hasn't enough capacity, then it might not run as well as one with abundance. So from a cycle point (speed?), zIIPs are not faster. Parwez Hamid -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
ZIIP and ZAAP processors always run at full speed, even when running on a sub-capacity box. One thing, among many, I don't know is how IBM implements sub-capacity. Slow the clock speed? Skip cycles? Chris Blaicher Technical Architect Syncsort, Inc. -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Parwez Sent: Sunday, March 3, 2019 6:05 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: CPU time and zIIP One regular misconception about (cycle time), irrespective of the type of processor the 'speed/cycle time' of ALL the processors is the SAME. CPs with CAPACITY setting of 7xx (Zxx in case of the 'BC' class system) are FULL capacity. Others are sub-capacity. So if the workload system hasn't enough capacity, then it might not run as well as one with abundance. So from a cycle point (speed?), zIIPs are not faster. Parwez Hamid -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
Or maybe it was ICFs. But, again, not quite 20 years. :-) Cheers, Martin Martin Packer zChampion, Systems Investigator & Performance Troubleshooter, IBM +44-7802-245-584 email: martin_pac...@uk.ibm.com Twitter / Facebook IDs: MartinPacker Blog: https://www.ibm.com/developerworks/mydeveloperworks/blogs/MartinPacker Podcast Series (With Marna Walle): https://developer.ibm.com/tv/mpt/or https://itunes.apple.com/gb/podcast/mainframe-performance-topics/id1127943573?mt=2 Youtube channel: https://www.youtube.com/channel/UCu_65HaYgksbF6Q8SQ4oOvA From: Peter Relson To: IBM-MAIN@LISTSERV.UA.EDU Date: 03/03/2019 15:03 Subject:Re: CPU time and zIIP Sent by:IBM Mainframe Discussion List >for maybe 20 years Not quite 20 years, but getting there . zAAPs were introduced in about 2004. Peter Relson z/OS Core Technology Design -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
>for maybe 20 years Not quite 20 years, but getting there . zAAPs were introduced in about 2004. Peter Relson z/OS Core Technology Design -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
One regular misconception about (cycle time), irrespective of the type of processor the 'speed/cycle time' of ALL the processors is the SAME. CPs with CAPACITY setting of 7xx (Zxx in case of the 'BC' class system) are FULL capacity. Others are sub-capacity. So if the workload system hasn't enough capacity, then it might not run as well as one with abundance. So from a cycle point (speed?), zIIPs are not faster. Parwez Hamid From: IBM Mainframe Discussion List on behalf of Mike Schwab Sent: 02 March 2019 21:07 To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: CPU time and zIIP Especially since zIIPs are faster but cost lest, and CPs are slower. One nice feature about reduced speed CPs is any delay waiting for resources are not counted toward your reduced speed for a CP. On Sat, Mar 2, 2019 at 2:49 PM Martin Packer wrote: > > Agreed. I - for maybe 20 years - have recommended reporting on each > processor pool separately. > > Cheers, Martin > > Martin Packer > > zChampion, Systems Investigator & Performance Troubleshooter, IBM > > +44-7802-245-584 > > email: martin_pac...@uk.ibm.com > > Twitter / Facebook IDs: MartinPacker > > Blog: > https://www.ibm.com/developerworks/mydeveloperworks/blogs/MartinPacker > > Podcast Series (With Marna Walle): https://developer.ibm.com/tv/mpt/or > > https://itunes.apple.com/gb/podcast/mainframe-performance-topics/id1127943573?mt=2 > > > Youtube channel: https://www.youtube.com/channel/UCu_65HaYgksbF6Q8SQ4oOvA > > > > From: Peter Relson > To: IBM-MAIN@LISTSERV.UA.EDU > Date: 02/03/2019 16:21 > Subject:Re: CPU time and zIIP > Sent by:IBM Mainframe Discussion List > > > > >is that true for CPU percent also? > > The original post and my answers were about SDSF. They did not show (to > the best of my eyesight) or discuss any percentage fields. > > Are you now asking a question about a percentage shown in RMF? I don't > pretend to know anything about what RMF displays, but it would likely not > be a good idea to try to lump together busy percentages pertaining to > standard CPs and zIIPs because the two types of processors are intended to > > have different characteristics in terms of how busy it is appropriate for > them to be. > > Peter Relson > z/OS Core Technology Design > > > -- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN > > > > > Unless stated otherwise above: > IBM United Kingdom Limited - Registered in England and Wales with number > 741598. > Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU > > -- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- Mike A Schwab, Springfield IL USA Where do Forest Rangers go to get away from it all? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
Especially since zIIPs are faster but cost lest, and CPs are slower. One nice feature about reduced speed CPs is any delay waiting for resources are not counted toward your reduced speed for a CP. On Sat, Mar 2, 2019 at 2:49 PM Martin Packer wrote: > > Agreed. I - for maybe 20 years - have recommended reporting on each > processor pool separately. > > Cheers, Martin > > Martin Packer > > zChampion, Systems Investigator & Performance Troubleshooter, IBM > > +44-7802-245-584 > > email: martin_pac...@uk.ibm.com > > Twitter / Facebook IDs: MartinPacker > > Blog: > https://www.ibm.com/developerworks/mydeveloperworks/blogs/MartinPacker > > Podcast Series (With Marna Walle): https://developer.ibm.com/tv/mpt/or > > https://itunes.apple.com/gb/podcast/mainframe-performance-topics/id1127943573?mt=2 > > > Youtube channel: https://www.youtube.com/channel/UCu_65HaYgksbF6Q8SQ4oOvA > > > > From: Peter Relson > To: IBM-MAIN@LISTSERV.UA.EDU > Date: 02/03/2019 16:21 > Subject:Re: CPU time and zIIP > Sent by:IBM Mainframe Discussion List > > > > >is that true for CPU percent also? > > The original post and my answers were about SDSF. They did not show (to > the best of my eyesight) or discuss any percentage fields. > > Are you now asking a question about a percentage shown in RMF? I don't > pretend to know anything about what RMF displays, but it would likely not > be a good idea to try to lump together busy percentages pertaining to > standard CPs and zIIPs because the two types of processors are intended to > > have different characteristics in terms of how busy it is appropriate for > them to be. > > Peter Relson > z/OS Core Technology Design > > > -- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN > > > > > Unless stated otherwise above: > IBM United Kingdom Limited - Registered in England and Wales with number > 741598. > Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU > > -- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- Mike A Schwab, Springfield IL USA Where do Forest Rangers go to get away from it all? -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
Agreed. I - for maybe 20 years - have recommended reporting on each processor pool separately. Cheers, Martin Martin Packer zChampion, Systems Investigator & Performance Troubleshooter, IBM +44-7802-245-584 email: martin_pac...@uk.ibm.com Twitter / Facebook IDs: MartinPacker Blog: https://www.ibm.com/developerworks/mydeveloperworks/blogs/MartinPacker Podcast Series (With Marna Walle): https://developer.ibm.com/tv/mpt/or https://itunes.apple.com/gb/podcast/mainframe-performance-topics/id1127943573?mt=2 Youtube channel: https://www.youtube.com/channel/UCu_65HaYgksbF6Q8SQ4oOvA From: Peter Relson To: IBM-MAIN@LISTSERV.UA.EDU Date: 02/03/2019 16:21 Subject:Re: CPU time and zIIP Sent by:IBM Mainframe Discussion List >is that true for CPU percent also? The original post and my answers were about SDSF. They did not show (to the best of my eyesight) or discuss any percentage fields. Are you now asking a question about a percentage shown in RMF? I don't pretend to know anything about what RMF displays, but it would likely not be a good idea to try to lump together busy percentages pertaining to standard CPs and zIIPs because the two types of processors are intended to have different characteristics in terms of how busy it is appropriate for them to be. Peter Relson z/OS Core Technology Design -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
>is that true for CPU percent also? The original post and my answers were about SDSF. They did not show (to the best of my eyesight) or discuss any percentage fields. Are you now asking a question about a percentage shown in RMF? I don't pretend to know anything about what RMF displays, but it would likely not be a good idea to try to lump together busy percentages pertaining to standard CPs and zIIPs because the two types of processors are intended to have different characteristics in terms of how busy it is appropriate for them to be. Peter Relson z/OS Core Technology Design -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
Carmen Vitullo - Original Message - From: "Peter Relson" To: IBM-MAIN@LISTSERV.UA.EDU Sent: Friday, March 1, 2019 7:15:51 AM Subject: Re: CPU time and zIIP I'm obviously still not understanding what you think is amiss. adding up the Gcpu time and Ziip time, to show all CP TIME but seeing that I misread, misunderstood the doc and your previous responses and from the statement belowas Gilda Radner would saynevermind :( for me I'd like to see SDSF's CPUtime to include all time, GCPU+IIP We have already said that it does. Peter Relson z/OS Core Technology Design is that true for CPU percent also? thanks for provided my the clue by four I needed to get this into my thick skull Carmen -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
I'm obviously still not understanding what you think is amiss. for me I'd like to see SDSF's CPUtime to include all time, GCPU+IIP We have already said that it does. Peter Relson z/OS Core Technology Design -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
I Probably wandering off the main topic of the OP's question, my initial response was according to SDSF ECPU% CPU usage consumed within the address space (RMF) to Brians query Hello, I'm trying to understand the CPU and ECPU times displayed on SDSF and the relation to zIIP processing time. For example, here is a CICS region running a Java web service. *CPU-Time ECPU-Time GCP-Time zIIP-Time zICP-Time zIIP-NTime* * 164.42 166.28 90.89 30.21 3.42 71.29* Here is a CICS transaction executing in the region to display various ASSB fields. CPU DISPLAY * *ASCBEJST. . . . . . : 162.20* *ASCBSRBT. . . . . . : 2.23* *ASSBASST. . . . . . : .00* *ASSBPHTM. . . . . . : 1.86* *CPU . . . . . . . . : 164.43 (ASCBEJST + ASCBSRBT + ASSBASST)* *ECPU. . . . . . . . : 166.29 (ASCBEJST + ASCBSRBT + ASSBPHTM)* *ASSB-TIME-ON-ZIIP . : 30.22* *ASSB-ZIIP-ENCT. . . : 1.60* *ASSB-ZIIP-PHTM. . . : 1.60* *ASSB-ASST-TIME-ON-CP: .00* The CPU-Time, ECPU-Time, and zIIP-time match (as close as humanly possible to switch 3270 emulators and press enter). I was under the assumption that ECPU also included time spent on zIIPs, but that must not be true. Is zIIP time included in CPU and ECPU? I'm thinking it is not. so..this . and it mutated from there trying to understand the CPU times displayed in SDSF, I personally don't use SDSF's ECPUtime, I'll use CPU% then if I see something out of the ordinary I'll check CPUtime and EXCP and SIO's use to be a good indicator of a looping program if CPU time is accumulating and EXCPs are zero or not accumulating . Pie in the sky if at all possible, for me I'd like to see SDSF's CPUtime to include all time, GCPU+IIP, it's nice the times and percentages are shown in a separate variable, I can see if my workload is really using the specialty engine, but SDSF does not report TCB and SRB times separately so why not show all CPUtime combined Carmen Vitullo - Original Message - From: "Peter Relson" To: IBM-MAIN@LISTSERV.UA.EDU Sent: Thursday, February 28, 2019 7:07:56 AM Subject: Re: CPU time and zIIP No more assumptions on my part, so what tools would show the correct time(s) for real time monitoring if not SDSF? RMF? Omegamon? And now that you throw Java in the mix or USS spanned tasks I don't think any real time monitor can account for all CPU time GCPU and zIIPTIME SDSF shows the correct time(s). What in any way leads you to think what is shown is not correct? SDSF shows what it shows. The data is what the data is. If you have a requirement for some other piece of data, then describe that in detail and we can tell you if it is available or not. In general the system keeps the data that if finds a need for, often (but not necessarily for only that reason) to make its way into SMF records. Peter Relson z/OS Core Technology Design -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
No more assumptions on my part, so what tools would show the correct time(s) for real time monitoring if not SDSF? RMF? Omegamon? And now that you throw Java in the mix or USS spanned tasks I don't think any real time monitor can account for all CPU time GCPU and zIIPTIME SDSF shows the correct time(s). What in any way leads you to think what is shown is not correct? SDSF shows what it shows. The data is what the data is. If you have a requirement for some other piece of data, then describe that in detail and we can tell you if it is available or not. In general the system keeps the data that if finds a need for, often (but not necessarily for only that reason) to make its way into SMF records. Peter Relson z/OS Core Technology Design -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
No more assumptions on my part, so what tools would show the correct time(s) for real time monitoring if not SDSF ? RMF? Omegamon? And now that you throw Java in the mix or USS spanned tasks I don't think any real time monitor can account for all CPU time GCPU and zIIPTIME I'm thinking none of the current tools can. Carmen Vitullo - Original Message - From: "Peter Relson" To: IBM-MAIN@LISTSERV.UA.EDU Sent: Wednesday, February 27, 2019 8:09:40 AM Subject: Re: CPU time and zIIP I see CPUtime as 4093.12, GCPUtime is 2187.06 and Ziiptime is 2.17, that's not adding up since all other times, other than accumulated time is zero There is no reason these should add up, although I don't have the information about where the "source" is for GCPUTime and "zIIP time", but I can guess. CPU time includes preemptable SRB time and non-preemptable SRB time. GCPU time likely does not (the field I am thinking of does not). zIIP time certainly does not include non-preemptable SRB time or any preemptable SRB time other than for enclave SRBs and even then only enclave SRBs that are running on zIIPs. And it will include task time for cases such as Java. I assumed since SDSF compartmentalized the zIIP, zAAP, and zICP times that they must not be included in the CPU-Time field. Like Carmen noted, when I add the GCP-Time, zIIP-Time, zAAP-Time, and zICP-Time it does not calculate out to be the same as the CPU-Time or ECPU-Time amount. Not a good assumption. These times were not "compartmentalized" by SDSF itself. SDSF displays using the "compartments" that the system uses for various reasons. It would be extremely surprising if the "addition" matched the CPU-Time or ECPU-Time. Peter Relson z/OS Core Technology Design -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
I see CPUtime as 4093.12, GCPUtime is 2187.06 and Ziiptime is 2.17, that's not adding up since all other times, other than accumulated time is zero There is no reason these should add up, although I don't have the information about where the "source" is for GCPUTime and "zIIP time", but I can guess. CPU time includes preemptable SRB time and non-preemptable SRB time. GCPU time likely does not (the field I am thinking of does not). zIIP time certainly does not include non-preemptable SRB time or any preemptable SRB time other than for enclave SRBs and even then only enclave SRBs that are running on zIIPs. And it will include task time for cases such as Java. I assumed since SDSF compartmentalized the zIIP, zAAP, and zICP times that they must not be included in the CPU-Time field. Like Carmen noted, when I add the GCP-Time, zIIP-Time, zAAP-Time, and zICP-Time it does not calculate out to be the same as the CPU-Time or ECPU-Time amount. Not a good assumption. These times were not "compartmentalized" by SDSF itself. SDSF displays using the "compartments" that the system uses for various reasons. It would be extremely surprising if the "addition" matched the CPU-Time or ECPU-Time. Peter Relson z/OS Core Technology Design -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
great info, thanks for the link Brian Carmen Vitullo - Original Message - From: "Brian Chapman" To: IBM-MAIN@LISTSERV.UA.EDU Sent: Tuesday, February 26, 2019 10:14:08 AM Subject: Re: CPU time and zIIP Thanks Peter. Here is the IBM document that I based my assumptions of the SDSF fields. *CPU-Time* Accumulated CPU time consumed by and on behalf of the address space, for the current job step, in seconds. SDSF obtains this value from RMF, as follows: ASCBEJST + ASCBSRBT + ASSBASST (source field R791TCPU) where *ASCBEJST* is elapsed job step time *ASCBSRBT* is accumulated SRB time *ASSBASST* is the CPU time consumed by preemptible class SRBs running on behalf of this address space, in milliseconds *ECPU-Time* Total CPU time consumed by and within the address space, for the current job step, in seconds. SDSF obtains this value from RMF, as follows: ASCBEJST + ASCBSRBT + ASSBPHTM (source field R791TCPC) where *ASCBEJST* is elapsed job step time *ASCBSRBT* is accumulated SRB time *ASSBPHTM* is the CPU time consumed by preemptible class SRBs running in this address space, in milliseconds (threads plus enclaves) https://www.ibm.com/support/knowledgecenter/en/SSLTBW_2.3.0/com.ibm.zosmfsdsf.jobresources.help.doc/izusfhpJobActiveSystemUse.html I assumed since SDSF compartmentalized the zIIP, zAAP, and zICP times that they must not be included in the CPU-Time field. Like Carmen noted, when I add the GCP-Time, zIIP-Time, zAAP-Time, and zICP-Time it does not calculate out to be the same as the CPU-Time or ECPU-Time amount. Thank you, Brian Chapman On Tue, Feb 26, 2019 at 9:21 AM Peter Relson wrote: > If the comment is correct that the SDSF display is using ASCBEJST, then > the statement > "ZIIP is not reported as part of CPU." > is not correct with respect to that display. > > ASCBEJST includes all time, whether standard CP or zIIP. > There are additional fields, such as ASSB_TIME_ON_CP, that do not include > zIIP. > > Fields in SMF records do use ASSB_TIME_ON_CP rather than ASCBEJST. For > those fields, the statement is correct. > > I wasn't sure in what way the OP concluded that zIIP was or was not > included when he wrote "must not be true" and "I'm thinking it is not". > > Peter Relson > z/OS Core Technology Design > > > -- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN > -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
Thanks Peter. Here is the IBM document that I based my assumptions of the SDSF fields. *CPU-Time* Accumulated CPU time consumed by and on behalf of the address space, for the current job step, in seconds. SDSF obtains this value from RMF, as follows: ASCBEJST + ASCBSRBT + ASSBASST (source field R791TCPU) where *ASCBEJST* is elapsed job step time *ASCBSRBT* is accumulated SRB time *ASSBASST* is the CPU time consumed by preemptible class SRBs running on behalf of this address space, in milliseconds *ECPU-Time* Total CPU time consumed by and within the address space, for the current job step, in seconds. SDSF obtains this value from RMF, as follows: ASCBEJST + ASCBSRBT + ASSBPHTM (source field R791TCPC) where *ASCBEJST* is elapsed job step time *ASCBSRBT* is accumulated SRB time *ASSBPHTM* is the CPU time consumed by preemptible class SRBs running in this address space, in milliseconds (threads plus enclaves) https://www.ibm.com/support/knowledgecenter/en/SSLTBW_2.3.0/com.ibm.zosmfsdsf.jobresources.help.doc/izusfhpJobActiveSystemUse.html I assumed since SDSF compartmentalized the zIIP, zAAP, and zICP times that they must not be included in the CPU-Time field. Like Carmen noted, when I add the GCP-Time, zIIP-Time, zAAP-Time, and zICP-Time it does not calculate out to be the same as the CPU-Time or ECPU-Time amount. Thank you, Brian Chapman On Tue, Feb 26, 2019 at 9:21 AM Peter Relson wrote: > If the comment is correct that the SDSF display is using ASCBEJST, then > the statement > "ZIIP is not reported as part of CPU." > is not correct with respect to that display. > > ASCBEJST includes all time, whether standard CP or zIIP. > There are additional fields, such as ASSB_TIME_ON_CP, that do not include > zIIP. > > Fields in SMF records do use ASSB_TIME_ON_CP rather than ASCBEJST. For > those fields, the statement is correct. > > I wasn't sure in what way the OP concluded that zIIP was or was not > included when he wrote "must not be true" and "I'm thinking it is not". > > Peter Relson > z/OS Core Technology Design > > > -- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN > -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
SDSF uses ERBSMFI for address space CPU-related information and processes the R791ELEM records mapped by ERBSMF79 in SYS1.MACLIB. The ECPU column in SDSF DA is populated from field R791TCPC. The comments in R791TCPC state the formula : ascbejst+ascbsrbt+(assbphtm-assbphtm_base) Rob Scott Rocket Software -Original Message- From: IBM Mainframe Discussion List On Behalf Of Peter Relson Sent: Tuesday, February 26, 2019 2:20 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: CPU time and zIIP If the comment is correct that the SDSF display is using ASCBEJST, then the statement "ZIIP is not reported as part of CPU." is not correct with respect to that display. ASCBEJST includes all time, whether standard CP or zIIP. There are additional fields, such as ASSB_TIME_ON_CP, that do not include zIIP. Fields in SMF records do use ASSB_TIME_ON_CP rather than ASCBEJST. For those fields, the statement is correct. I wasn't sure in what way the OP concluded that zIIP was or was not included when he wrote "must not be true" and "I'm thinking it is not". Peter Relson z/OS Core Technology Design -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN Rocket Software, Inc. and subsidiaries ■ 77 Fourth Avenue, Waltham MA 02451 ■ Main Office Toll Free Number: +1 855.577.4323 Contact Customer Support: https://my.rocketsoftware.com/RocketCommunity/RCEmailSupport Unsubscribe from Marketing Messages/Manage Your Subscription Preferences - http://www.rocketsoftware.com/manage-your-email-preferences Privacy Policy - http://www.rocketsoftware.com/company/legal/privacy-policy This communication and any attachments may contain confidential information of Rocket Software, Inc. All unauthorized use, disclosure or distribution is prohibited. If you are not the intended recipient, please notify Rocket Software immediately and destroy all copies of this communication. Thank you. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
I think the OP and prolly most of us non developers make a bad assumption based on what SDSF's help and user guide provides, which is very little. checking the alternate display on SDSF for one of my DB2 subsystems I see CPUtime as 4093.12, GCPUtime is 2187.06 and Ziiptime is 2.17, that's not adding up since all other times, other than accumulated time is zero Carmen Vitullo - Original Message - From: "Peter Relson" To: IBM-MAIN@LISTSERV.UA.EDU Sent: Tuesday, February 26, 2019 8:20:00 AM Subject: CPU time and zIIP If the comment is correct that the SDSF display is using ASCBEJST, then the statement "ZIIP is not reported as part of CPU." is not correct with respect to that display. ASCBEJST includes all time, whether standard CP or zIIP. There are additional fields, such as ASSB_TIME_ON_CP, that do not include zIIP. Fields in SMF records do use ASSB_TIME_ON_CP rather than ASCBEJST. For those fields, the statement is correct. I wasn't sure in what way the OP concluded that zIIP was or was not included when he wrote "must not be true" and "I'm thinking it is not". Peter Relson z/OS Core Technology Design -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
CPU time and zIIP
If the comment is correct that the SDSF display is using ASCBEJST, then the statement "ZIIP is not reported as part of CPU." is not correct with respect to that display. ASCBEJST includes all time, whether standard CP or zIIP. There are additional fields, such as ASSB_TIME_ON_CP, that do not include zIIP. Fields in SMF records do use ASSB_TIME_ON_CP rather than ASCBEJST. For those fields, the statement is correct. I wasn't sure in what way the OP concluded that zIIP was or was not included when he wrote "must not be true" and "I'm thinking it is not". Peter Relson z/OS Core Technology Design -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
according to SDSF ECPU% CPU usage consumed within the address space (RMF) Carmen Vitullo - Original Message - From: "Brian Chapman" To: IBM-MAIN@LISTSERV.UA.EDU Sent: Monday, February 25, 2019 12:52:35 PM Subject: CPU time and zIIP Hello, I'm trying to understand the CPU and ECPU times displayed on SDSF and the relation to zIIP processing time. For example, here is a CICS region running a Java web service. *CPU-Time ECPU-Time GCP-Time zIIP-Time zICP-Time zIIP-NTime* * 164.42 166.28 90.89 30.21 3.42 71.29* Here is a CICS transaction executing in the region to display various ASSB fields. CPU DISPLAY * *ASCBEJST. . . . . . : 162.20* *ASCBSRBT. . . . . . : 2.23* *ASSBASST. . . . . . : .00* *ASSBPHTM. . . . . . : 1.86* *CPU . . . . . . . . : 164.43 (ASCBEJST + ASCBSRBT + ASSBASST)* *ECPU. . . . . . . . : 166.29 (ASCBEJST + ASCBSRBT + ASSBPHTM)* *ASSB-TIME-ON-ZIIP . : 30.22* *ASSB-ZIIP-ENCT. . . : 1.60* *ASSB-ZIIP-PHTM. . . : 1.60* *ASSB-ASST-TIME-ON-CP: .00* The CPU-Time, ECPU-Time, and zIIP-time match (as close as humanly possible to switch 3270 emulators and press enter). I was under the assumption that ECPU also included time spent on zIIPs, but that must not be true. Is zIIP time included in CPU and ECPU? I'm thinking it is not. Thank you, Brian Chapman -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: CPU time and zIIP
ZIIP is not reported as part of CPU. Chris Blaicher Technical Architect Syncsort, Inc. -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Brian Chapman Sent: Monday, February 25, 2019 1:53 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: CPU time and zIIP Hello, I'm trying to understand the CPU and ECPU times displayed on SDSF and the relation to zIIP processing time. For example, here is a CICS region running a Java web service. *CPU-Time ECPU-Time GCP-Time zIIP-Time zICP-Time zIIP-NTime* * 164.42166.2890.89 30.21 3.42 71.29* Here is a CICS transaction executing in the region to display various ASSB fields. CPU DISPLAY * *ASCBEJST. . . . . . : 162.20* *ASCBSRBT. . . . . . :2.23* *ASSBASST. . . . . . : .00* *ASSBPHTM. . . . . . :1.86* *CPU . . . . . . . . : 164.43 (ASCBEJST + ASCBSRBT + ASSBASST)* *ECPU. . . . . . . . : 166.29 (ASCBEJST + ASCBSRBT + ASSBPHTM)* *ASSB-TIME-ON-ZIIP . : 30.22* *ASSB-ZIIP-ENCT. . . :1.60* *ASSB-ZIIP-PHTM. . . :1.60* *ASSB-ASST-TIME-ON-CP: .00* The CPU-Time, ECPU-Time, and zIIP-time match (as close as humanly possible to switch 3270 emulators and press enter). I was under the assumption that ECPU also included time spent on zIIPs, but that must not be true. Is zIIP time included in CPU and ECPU? I'm thinking it is not. Thank you, Brian Chapman -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
CPU time and zIIP
Hello, I'm trying to understand the CPU and ECPU times displayed on SDSF and the relation to zIIP processing time. For example, here is a CICS region running a Java web service. *CPU-Time ECPU-Time GCP-Time zIIP-Time zICP-Time zIIP-NTime* * 164.42166.2890.89 30.21 3.42 71.29* Here is a CICS transaction executing in the region to display various ASSB fields. CPU DISPLAY * *ASCBEJST. . . . . . : 162.20* *ASCBSRBT. . . . . . :2.23* *ASSBASST. . . . . . : .00* *ASSBPHTM. . . . . . :1.86* *CPU . . . . . . . . : 164.43 (ASCBEJST + ASCBSRBT + ASSBASST)* *ECPU. . . . . . . . : 166.29 (ASCBEJST + ASCBSRBT + ASSBPHTM)* *ASSB-TIME-ON-ZIIP . : 30.22* *ASSB-ZIIP-ENCT. . . :1.60* *ASSB-ZIIP-PHTM. . . :1.60* *ASSB-ASST-TIME-ON-CP: .00* The CPU-Time, ECPU-Time, and zIIP-time match (as close as humanly possible to switch 3270 emulators and press enter). I was under the assumption that ECPU also included time spent on zIIPs, but that must not be true. Is zIIP time included in CPU and ECPU? I'm thinking it is not. Thank you, Brian Chapman -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
