Re: Storage paradigm [was: RE: Data volumes]
l...@garlic.com (Anne Lynn Wheeler) writes: In the transition from MVT to OS/VS2 (aka virtual memory), the same problem showed up. The original implementation involved putting a little bit of code to create 16mbyte virtual address space for MVT, but the major effort was hacking CCWTRANS (from CP67) into the side of EXCP processing (EXCP had the same problem with access methods creating channel programs in the application virtual address space ... as CP67 did with virtual machine channel programs). Old reference by somebody in POK that was in the middle of the transition to OS/VS2 and virtual memory ... includes reference that OS/VS2 release 2 (MVS) was on glide path to OS/VS2 release 3 (FS) re: http://www.garlic.com/~lynn/2013h.html#44 Why does IBM keep saying things like this http://www.garlic.com/~lynn/2013h.html#45 Storage paradigm [was: RE: Data volumes] http://www.garlic.com/~lynn/2013h.html#47 Storage paradigm [was: RE: Data volumes] just finished scanning my copy of cp67 PLM (although it was missing a couple pages) ... and uploading to bitsavers ... it should be showing up shortly in http://bitsavers.org/pdf/ibm/360/cp67 CP67 CCWTRANS (which was hacked into EXCP to perform the channel program virtual-real translation for OS/VS2) is described starting on page 35 (physical page 45 in the PDF file). -- virtualization experience starting Jan1968, online at home since Mar1970 -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Data volumes
At 08:18 -0500 on 06/11/2013, Paul Gilmartin wrote about Re: Data volumes: Would that be 54 GB? You have a thumb (flash?) drive bigger than that? My preferred Thumb Drive size are 64GB. They are a few dollars cheaper than 2 32GB dives and hold more data since there is only one directory not two. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Storage paradigm [was: RE: Data volumes]
Perhaps you are keeping bad company. While humans are not perfect, there are methods to improve code reliability. In 30+ years, I've worked for 6 companies. Gee, they all must be bad company! - Ted MacNEIL eamacn...@yahoo.ca Twitter: @TedMacNEIL -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Storage paradigm [was: RE: Data volumes]
W dniu 2013-06-11 08:58, Ted MacNEIL pisze: Perhaps you are keeping bad company. While humans are not perfect, there are methods to improve code reliability. In 30+ years, I've worked for 6 companies. Gee, they all must be bad company! Are we still talking about possible bug in the aplication which can cause enormous disk space consumption? If so, I can say I don't know how LUW systems could manage it without SPACE=(), but they do, definitely. -- Radoslaw Skorupka Lodz, Poland -- Treść tej wiadomości może zawierać informacje prawnie chronione Banku przeznaczone wyłącznie do użytku służbowego adresata. Odbiorcą może być jedynie jej adresat z wyłączeniem dostępu osób trzecich. Jeżeli nie jesteś adresatem niniejszej wiadomości lub pracownikiem upoważnionym do jej przekazania adresatowi, informujemy, że jej rozpowszechnianie, kopiowanie, rozprowadzanie lub inne działanie o podobnym charakterze jest prawnie zabronione i może być karalne. Jeżeli otrzymałeś tę wiadomość omyłkowo, prosimy niezwłocznie zawiadomić nadawcę wysyłając odpowiedź oraz trwale usunąć tę wiadomość włączając w to wszelkie jej kopie wydrukowane lub zapisane na dysku. This e-mail may contain legally privileged information of the Bank and is intended solely for business use of the addressee. This e-mail may only be received by the addressee and may not be disclosed to any third parties. If you are not the intended addressee of this e-mail or the employee authorised to forward it to the addressee, be advised that any dissemination, copying, distribution or any other similar activity is legally prohibited and may be punishable. If you received this e-mail by mistake please advise the sender immediately by using the reply facility in your e-mail software and delete permanently this e-mail including any copies of it either printed or saved to hard drive. BRE Bank SA, 00-950 Warszawa, ul. Senatorska 18, tel. +48 (22) 829 00 00, fax +48 (22) 829 00 33, www.brebank.pl, e-mail: i...@brebank.pl Sąd Rejonowy dla m. st. Warszawy XII Wydział Gospodarczy Krajowego Rejestru Sądowego, nr rejestru przedsiębiorców KRS 025237, NIP: 526-021-50-88. Według stanu na dzień 01.01.2013 r. kapitał zakładowy BRE Banku SA (w całości wpłacony) wynosi 168.555.904 złotych. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Storage paradigm [was: RE: Data volumes]
There is no need to be so afraid of runaway disk space consumption, when we don't control similar problems that precisely in other areas: spoolspace utilization, job execution time (/*JOBPARM T=), GBs written to tape, etc. etc. In those areas, the control is to let people do their thing without prior specification of prediction and only take action if certain limits are exceeded. Why not tread Dasd the same: let people specify Dataclass and skip SPACE (we can now already) and set limits on the size of the dataset per Dataclass. Like JES: Warn or abend the job if the size is exceeded, control who can use different Dataclasses etc. Kees. -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Gerhard Postpischil Sent: Monday, June 10, 2013 19:32 To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: Storage paradigm [was: RE: Data volumes] On 6/10/2013 11:38 AM, Farley, Peter x23353 wrote: Rant Like a few others on this list, I have often gritted my teeth at the necessity to estimate disk storage quantities that vary widely over time in a fixed manner (i.e., SPACE in JCL) when the true need is just to match output volume to input volume each day. If it's that predictable, it's trivial to write code to produce an estimated output volume from input, and tailor and submit the appropriate JCL. So that's a non-issue. EAV or not EAV, guaranteed space or not, candidate volumes, striped or not striped, compressed or not compressed - all of that baggage is clearly non-optimal for getting the job done in a timely manner. Why should allocating a simple sequential file require a team of Storage Administration experts to accomplish effectively? /Rant There is no theoretical solution. On any system running jobs, it is possible for one job to monopolize available space, requiring other jobs to wait forever or be terminated. Even on a single job system that job may exhaust space. Requiring a space specification may be a PITA, but it guarantees that a started job will finish (subject to other constraints). And the SA experts, especially for sequential files, can be avoided with simple estimator programs. This seems to be more of a religious war than a practical discussion. Gerhard Postpischil Bradford, Vermont -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN For information, services and offers, please visit our web site: http://www.klm.com. This e-mail and any attachment may contain confidential and privileged material intended for the addressee only. If you are not the addressee, you are notified that no part of the e-mail or any attachment may be disclosed, copied or distributed, and that any other action related to this e-mail or attachment is strictly prohibited, and may be unlawful. If you have received this e-mail by error, please notify the sender immediately by return e-mail, and delete this message. Koninklijke Luchtvaart Maatschappij NV (KLM), its subsidiaries and/or its employees shall not be liable for the incorrect or incomplete transmission of this e-mail or any attachments, nor responsible for any delay in receipt. Koninklijke Luchtvaart Maatschappij N.V. (also known as KLM Royal Dutch Airlines) is registered in Amstelveen, The Netherlands, with registered number 33014286 -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Storage paradigm [was: RE: Data volumes]
Hear, hear! Regards Thomas Berg Thomas Berg Specialist z/OS\RQM\IT Delivery SWEDBANK AB (Publ) -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Farley, Peter x23353 Sent: Monday, June 10, 2013 5:38 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Storage paradigm [was: RE: Data volumes] Rant Like a few others on this list, I have often gritted my teeth at the necessity to estimate disk storage quantities that vary widely over time in a fixed manner (i.e., SPACE in JCL) when the true need is just to match output volume to input volume each day. Why is it that IBM (and organizations that use their mainframe systems) so vigorously resist a conversion off of the ECKD standard? (Yes, I know it's all about conversion cost, but in the larger picture that is a red herring.) Not that I'm likely to see such a transition in my lifetime, but in this dawning time of soi-disant big data, perhaps it is past time to change the storage paradigm entirely, not from ECKD to FBA but to transition instead to something like the Multics model where every object in the system (whether in memory or on external storage, whether data or program) has an address, and all addresses are unique. Let the storage subsystem decide how to optimally position and aggregate the various parts of objects, and how to organize them for best performance. Such decisions should not require human guesstimate input to be optimal, or nearly so. Characteristics of application access are far more critical specifications than mere size. The ability to specify just the desired application access characteristics (random, sequential, growing, shrinking, response-time-critical, etc.) should be necessary and sufficient. EAV or not EAV, guaranteed space or not, candidate volumes, striped or not striped, compressed or not compressed - all of that baggage is clearly non-optimal for getting the job done in a timely manner. Why should allocating a simple sequential file require a team of Storage Administration experts to accomplish effectively? /Rant Peter -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Ed Jaffe Sent: Sunday, June 09, 2013 10:47 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: Data volumes On 6/9/2013 7:12 AM, Scott Ford wrote: We need bigger dasd ...ouch The largest 3390 volumes in our tiny shop hold 3,940,020 tracks or 262,668 cylinders. That is the maximum size supported by the back-level DASD we are running. Newer DASD hardware can support volumes up to 1TB in size. I assume nearly all zEC12 and z196 customers are capable of exploiting these large sizes. But, do they? I spent three years dealing with, and eventually helping IBM to solve (via OA40210 - HIPER, DATALOSS), a serious EAV bug that should have been seen in most every shop in the world that uses the DFSMSdss CONSOLIDATE function (with or without DEFRAG). The experience was a real eye-opener for me and I concluded that almost nobody is using EAV! Why not? Personally, I would find it embarrassing if the Corsair thumb drive in my pocket held more data than our largest mainframe volumes. But, that's just me... -- This message and any attachments are intended only for the use of the addressee and may contain information that is privileged and confidential. If the reader of the message is not the intended recipient or an authorized representative of the intended recipient, you are hereby notified that any dissemination of this communication is strictly prohibited. If you have received this communication in error, please notify us immediately by e-mail and delete the message and any attachments from your system. -- 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: Storage paradigm [was: RE: Data volumes]
Gerhard Postpischil wrote: What planet are you from? Sol 3 Interesting that you refer to that filthy big blue polluted ironball with warring citizens. The last time I checked, we're on Tatooine in a galaxy far far away or so I think... ;-) Ted MacNEIL wrote: Programmers seem able to test everything except that one condition that will break in Production Murphy's law. You design/build a job, program, etc and when going live / production, something crash. Fix it and live with it. Sometimes when going Production, it is actually the LAST major test to iron out any last wrinkles. Groete / Greetings Elardus Engelbrecht Joke of the day (Sol 3 of course): One day a teacher was talking about marriage in class. Teacher : What kind of wife would you like Johnny? Johnny : I would want a wife like the moon. Teacher : Wow! What a choice... Do you want her to be beautiful and calm like the moon? Johnny : No, I want her to arrive at night and disappear in the morning... Give that boy a Bells!!! -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Data volumes
W dniu 2013-06-11 15:18, Paul Gilmartin pisze: FSVO broken? I suppose a GB of random bits could be deemed a broken VBS dataset. For that there's no plausible repair. How do VBS data sets get broken? Is a VB data set immune to such depravities, or do the errors simply pass undetected? We are talking about SMF. It's a kind of log, a bunch of records. While it's very good (sometimes crucial) to have all of the records, sometimes we lose one of the billions of records and it is still better to have all of them-1 (or few more) than nothing. VB or FB is not immune, but there is kind f failure which is specific to VBS, because only VBS can have lost segments. So, only VBS can be fixed (by substraction incomplete record). Last, but not least: GB of random bits is not broken VBS file, as bunch of random letters is very unlikely to form a Hamlet. -- Radoslaw Skorupka Lodz, Poland -- Treść tej wiadomości może zawierać informacje prawnie chronione Banku przeznaczone wyłącznie do użytku służbowego adresata. Odbiorcą może być jedynie jej adresat z wyłączeniem dostępu osób trzecich. Jeżeli nie jesteś adresatem niniejszej wiadomości lub pracownikiem upoważnionym do jej przekazania adresatowi, informujemy, że jej rozpowszechnianie, kopiowanie, rozprowadzanie lub inne działanie o podobnym charakterze jest prawnie zabronione i może być karalne. Jeżeli otrzymałeś tę wiadomość omyłkowo, prosimy niezwłocznie zawiadomić nadawcę wysyłając odpowiedź oraz trwale usunąć tę wiadomość włączając w to wszelkie jej kopie wydrukowane lub zapisane na dysku. This e-mail may contain legally privileged information of the Bank and is intended solely for business use of the addressee. This e-mail may only be received by the addressee and may not be disclosed to any third parties. If you are not the intended addressee of this e-mail or the employee authorised to forward it to the addressee, be advised that any dissemination, copying, distribution or any other similar activity is legally prohibited and may be punishable. If you received this e-mail by mistake please advise the sender immediately by using the reply facility in your e-mail software and delete permanently this e-mail including any copies of it either printed or saved to hard drive. BRE Bank SA, 00-950 Warszawa, ul. Senatorska 18, tel. +48 (22) 829 00 00, fax +48 (22) 829 00 33, www.brebank.pl, e-mail: i...@brebank.pl Sąd Rejonowy dla m. st. Warszawy XII Wydział Gospodarczy Krajowego Rejestru Sądowego, nr rejestru przedsiębiorców KRS 025237, NIP: 526-021-50-88. Według stanu na dzień 01.01.2013 r. kapitał zakładowy BRE Banku SA (w całości wpłacony) wynosi 168.555.904 złotych. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Data volumes
On Jun 11, 2013, at 8:18 AM, Paul Gilmartin paulgboul...@aim.com wrote: 2**18 + 2**9 + 12. A peculiar number. I wonder how they chose that? For what it's worth, that's 236 * 1113, the latter being the size of a 3390 mod 1. I don't know why they went with 236 (2**2 * 59) though. -- Curtis Pew (c@its.utexas.edu) ITS Systems Core The University of Texas at Austin -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Storage paradigm [was: RE: Data volumes]
On 06/10/2013 01:57 PM, Ted MacNEIL wrote: As to run-away programs, they should be thoroughly checked on a test system before going into production; a run-away in production should be so rare as to be immaterial. What planet are you from? Programmers seem able to test everything except that one condition that will break in Production - Ted MacNEIL eamacn...@yahoo.ca Twitter: @TedMacNEIL Another way to look at it is that there are usually many, many more users than application testers. A clever, curious, or just plain inept user can always find many more weird combinations of input to an application than can be tested, sometimes nonsensical ones that would just never occur to a tester to try; and at least a few of these may give highly undesirable results. Consider z/OS itself as a case in point. One would hope this is an example of extensively tested software, yet every release always has a number of fix PTFs, some of which address problems that were highly fatal to one or more installations. -- Joel C. Ewing,Bentonville, AR jcew...@acm.org -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Data volumes
Paul Gilmartin wrote: How do VBS data sets get broken? From my sour experiences on this planet I will try to explain. In my SMF jobs, years ago, I get nasty 613 abends due to channel error (example message from my documentation) IEC147I 613-1C,IFG0194K,SMFDMP,TAPE,SYSUT2,0C30,... IEE763I NAME= IFG0194K CODE= 6131C IEC147I LOGREC ID = IEE763I NAME= CBRLLACS CODE= 140366 Now, if it does go undetected and you MOD (next day) after that tape (RC=00), all those MODding are just a waste of time and resources, since you can only read your SMF data up to the point on the tape where that abend took place. So your SMF tape layout is something like this: volser label file label rec 1 . last full intact rec n half written record - point of 613 abend --- unreadable by IFASMFDP MODded rec 1 --- Can that program mentioned by Barry Merrill read it? ... MODded rec last EOF + EOV To recover, I read all SMF data of that tape immediately up to the very last readable record to ANOTHER tape. Then I MOD the rest of my SMF data still waiting to be written to that new tape. When RC=00, I copy everything back to the broken SMF data on my tape to reuse it and make it ready for next day SMF job. And make a new backup of course. I got absolutel weary and p*ssed off about those channel errors. Hope this explain those 'broken' VBS datasets on tape. I'm not sure how to explain such broken datasets on DASD. Groete / Greetings Elardus Engelbrecht -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Data volumes
I wrote a FIXVBS program which read each segment separately (as a VB record) and reconstructed the logical record. If there was a problem during reconstruction, I wrote an error message and started with the next begin segment. On Tue, Jun 11, 2013 at 8:56 AM, Elardus Engelbrecht elardus.engelbre...@sita.co.za wrote: Paul Gilmartin wrote: How do VBS data sets get broken? From my sour experiences on this planet I will try to explain. In my SMF jobs, years ago, I get nasty 613 abends due to channel error (example message from my documentation) IEC147I 613-1C,IFG0194K,SMFDMP,TAPE,SYSUT2,0C30,... IEE763I NAME= IFG0194K CODE= 6131C IEC147I LOGREC ID = IEE763I NAME= CBRLLACS CODE= 140366 Now, if it does go undetected and you MOD (next day) after that tape (RC=00), all those MODding are just a waste of time and resources, since you can only read your SMF data up to the point on the tape where that abend took place. So your SMF tape layout is something like this: volser label file label rec 1 . last full intact rec n half written record - point of 613 abend --- unreadable by IFASMFDP MODded rec 1 --- Can that program mentioned by Barry Merrill read it? ... MODded rec last EOF + EOV To recover, I read all SMF data of that tape immediately up to the very last readable record to ANOTHER tape. Then I MOD the rest of my SMF data still waiting to be written to that new tape. When RC=00, I copy everything back to the broken SMF data on my tape to reuse it and make it ready for next day SMF job. And make a new backup of course. I got absolutel weary and p*ssed off about those channel errors. Hope this explain those 'broken' VBS datasets on tape. I'm not sure how to explain such broken datasets on DASD. Groete / Greetings Elardus Engelbrecht -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- This is a test of the Emergency Broadcast System. If this had been an actual emergency, do you really think we'd stick around to tell you? Maranatha! John McKown -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Storage paradigm [was: RE: Data volumes]
On 6/11/2013 5:41 AM, Elardus Engelbrecht wrote: Interesting that you refer to that filthy big blue polluted ironball with warring citizens. The last time I checked, we're on Tatooine in a galaxy far far away or so I think... ;-) Nice display of dry wit G Gerhard Postpischil Bradford, Vermont -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Storage paradigm [was: RE: Data volumes]
In CAAJSdjhdFqcpkSUKJ9Uu+DSJh1VwwJdFT2VSHwC3P9=79nc...@mail.gmail.com, on 06/10/2013 at 11:45 AM, John McKown john.archie.mck...@gmail.com said: LUW works similar to z/OS UNIX file systems. I.e. there is a file system which is formatted using some utility (mkfs in the Linux/UNIX world, format in Windows). This sets up all the internals. In today's LUW, it is usually possible for a single file to be as big as the file system upon which it resides. But no bigger. There is nothing like a multi file system file (which would vaguely like a multivolume data set). I don't know about windoze, but don't LVM and EVMS allow file systems to span devices? -- Shmuel (Seymour J.) Metz, SysProg and JOAT Atid/2http://patriot.net/~shmuel We don't care. We don't have to care, we're Congress. (S877: The Shut up and Eat Your spam act of 2003) -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Storage paradigm [was: RE: Data volumes]
In 51b60a0b.7030...@valley.net, on 06/10/2013 at 01:16 PM, Gerhard Postpischil gerh...@valley.net said: Technically the easiest to implement would be adding a new device type, thus keeping (E)CKD completely distinct from FBA. The new type could be supported by VSAM/AMS only (and JCL, SVC 99, etc.) without impacting other programs. (I would hope there are no programs out there using TM UCBTBYT3 rather than CLI?) If IBM could implement a VSAM ESDS reverse compatibility interface (RCI) for other systems, why not for MVS? -- Shmuel (Seymour J.) Metz, SysProg and JOAT Atid/2http://patriot.net/~shmuel We don't care. We don't have to care, we're Congress. (S877: The Shut up and Eat Your spam act of 2003) -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Storage paradigm [was: RE: Data volumes]
In 985915eee6984740ae93f8495c624c6c23194bd...@jscpcwexmaa1.bsg.ad.adp.com, on 06/10/2013 at 02:46 PM, Farley, Peter x23353 peter.far...@broadridge.com said: There *are* non-theoretical solutions to runaway file output. The *ix system model of using disk quotas per user makes it entirely possible to imagine z/OS application users with reasonable disk quotas specific to the application (i.e., not by job but by suite of jobs). Mommy, make it go away. An out of control job would kill unrelated work. Per file quotas might be workable, but would lead to the same types of outcry as space specification: I don't know how much output the job will create. -- Shmuel (Seymour J.) Metz, SysProg and JOAT Atid/2http://patriot.net/~shmuel We don't care. We don't have to care, we're Congress. (S877: The Shut up and Eat Your spam act of 2003) -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Storage paradigm [was: RE: Data volumes]
In 985915eee6984740ae93f8495c624c6c23194bd...@jscpcwexmaa1.bsg.ad.adp.com, on 06/10/2013 at 11:38 AM, Farley, Peter x23353 peter.far...@broadridge.com said: Why is it that IBM (and organizations that use their mainframe systems) so vigorously resist a conversion off of the ECKD standard? Before asking why, ask whether. I have seen no evidence that anybody but IBM resists such a change. perhaps it is past time to change the storage paradigm entirely, not from ECKD to FBA but to transition instead to something like the Multics model where every object in the system (whether in memory or on external storage, whether data or program) has an address, and all addresses are unique. That's not the Multics model. The Multic model is that segment numbers are dynamically assigned as needed, and that in general two processes will use different numbers for the same segment. IBM had something similar in TSS, but abandoned it. -- Shmuel (Seymour J.) Metz, SysProg and JOAT Atid/2http://patriot.net/~shmuel We don't care. We don't have to care, we're Congress. (S877: The Shut up and Eat Your spam act of 2003) -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Storage paradigm [was: RE: Data volumes]
On 6/11/2013 2:58 AM, Ted MacNEIL wrote: In 30+ years, I've worked for 6 companies. Gee, they all must be bad company! I'll take your word for it. In 40+ years I worked for 8 companies, two of which were ISVs, two service bureaus, and the rest were contract software providers (mostly, but not all for government agencies). Many had strict procedures for how code was inspected and tested; one went so far as to reduce any abending production program back to test status, even when the abend was due to a hardware failure. In all that time, I saw programs that evinced run-away space consumption, but they were all caught during testing. Gerhard Postpischil Bradford, Vermont -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Storage paradigm [was: RE: Data volumes]
LVM requires an LV (Logical Volume) to be in a single VG (Volume Group). A VG is composed of one or more PVs (Physical Volumes). A PV is basically a specially formatted disk partition (which can be the entire disk or subdivision). A single file system must reside on a single LV. Which is in a VG. Which can span multiple PVs. And so, in a somewhat indirect manner, a file system can span multiple physical volumes. http://www.web-manual.net/linux-3/logical-volume-manager-in-linux/ On Tue, Jun 11, 2013 at 8:28 AM, Shmuel Metz (Seymour J.) shmuel+...@patriot.net wrote: In CAAJSdjhdFqcpkSUKJ9Uu+DSJh1VwwJdFT2VSHwC3P9=79nc...@mail.gmail.com, on 06/10/2013 at 11:45 AM, John McKown john.archie.mck...@gmail.com said: LUW works similar to z/OS UNIX file systems. I.e. there is a file system which is formatted using some utility (mkfs in the Linux/UNIX world, format in Windows). This sets up all the internals. In today's LUW, it is usually possible for a single file to be as big as the file system upon which it resides. But no bigger. There is nothing like a multi file system file (which would vaguely like a multivolume data set). I don't know about windoze, but don't LVM and EVMS allow file systems to span devices? -- Shmuel (Seymour J.) Metz, SysProg and JOAT Atid/2http://patriot.net/~shmuel We don't care. We don't have to care, we're Congress. (S877: The Shut up and Eat Your spam act of 2003) -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- This is a test of the Emergency Broadcast System. If this had been an actual emergency, do you really think we'd stick around to tell you? Maranatha! John McKown -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Storage paradigm [was: RE: Data volumes]
Yes, most people basically want to say: I don't want to have to concern myself with how much space and time it takes to do my work. I just want it to work. Oh, and I need it ASAP. And it must not require too much thought. That is, I need to be able to use it on a Monday morning, before I have had my first cup of coffee. Oh, and it needs to be free so that it doesn't impact my budget. That's so I can be under budget and get a raise or bonus. cynic type=old/ On Tue, Jun 11, 2013 at 8:37 AM, Shmuel Metz (Seymour J.) shmuel+...@patriot.net wrote: In 985915eee6984740ae93f8495c624c6c23194bd...@jscpcwexmaa1.bsg.ad.adp.com, on 06/10/2013 at 02:46 PM, Farley, Peter x23353 peter.far...@broadridge.com said: There *are* non-theoretical solutions to runaway file output. The *ix system model of using disk quotas per user makes it entirely possible to imagine z/OS application users with reasonable disk quotas specific to the application (i.e., not by job but by suite of jobs). Mommy, make it go away. An out of control job would kill unrelated work. Per file quotas might be workable, but would lead to the same types of outcry as space specification: I don't know how much output the job will create. -- Shmuel (Seymour J.) Metz, SysProg and JOAT Atid/2http://patriot.net/~shmuel We don't care. We don't have to care, we're Congress. (S877: The Shut up and Eat Your spam act of 2003) -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- This is a test of the Emergency Broadcast System. If this had been an actual emergency, do you really think we'd stick around to tell you? Maranatha! John McKown -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Storage paradigm [was: RE: Data volumes]
-Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Shmuel Metz (Seymour J.) Sent: Tuesday, June 11, 2013 3:37 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: Storage paradigm [was: RE: Data volumes] In 985915eee6984740ae93f8495c624c6c23194bd...@jscpcwexmaa1.bsg.ad.adp.com , on 06/10/2013 at 02:46 PM, Farley, Peter x23353 peter.far...@broadridge.com said: There *are* non-theoretical solutions to runaway file output. The *ix system model of using disk quotas per user makes it entirely possible to imagine z/OS application users with reasonable disk quotas specific to the application (i.e., not by job but by suite of jobs). Mommy, make it go away. An out of control job would kill unrelated work. Per file quotas might be workable, but would lead to the same types of outcry as space specification: I don't know how much output the job will create. No. Regards Thomas Berg Thomas Berg Specialist z/OS\RQM\IT Delivery SWEDBANK AB (Publ) -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Storage paradigm [was: RE: Data volumes]
shmuel+...@patriot.net (Shmuel Metz , Seymour J.) writes: That's not the Multics model. The Multic model is that segment numbers are dynamically assigned as needed, and that in general two processes will use different numbers for the same segment. IBM had something similar in TSS, but abandoned it. some of the people from CTSS went to the 5th flr and did Project Mac multics. other of the people went to the science center on the 4th flr and did virtual machines, online computing, the internal network ... GML was invented at the science center in 1969, as well as lots of performance monitoring and modeling stuff ... some of which evolves into capacity planning ... misc. past posts mentioning 545 tech sq http://www.garlic.com/~lynn/subtopic.html#545tech supposedly 360/67 (360/65 with virtual memory) was to be ibm's candidate for project mac ... but ge won the bid instead ... and multics was done on ge645. melinda's history has lots of details ... can be found here: http://www.leeandmelindavarian.com/Melinda/ tss/360 was then going to the official operating system ... including virtual memory and single-level-store (storage mapped as virtual memory ... 360/67 had both 24bit and 32bit virtual addressing modes). at one point tss/360 is claimed to have something like 1200 people at the time the science people had 12 people on cp40/cms (before being able to get 360/67, they modify 360/40 with virtual memory hardwre, later when they were able to get 360/67, cp40/cms morphs into cp67/cms ... and later into vm370/cms). old user group presentationon cp40 http://www.garlic.com/~lynn/cp40seas1982.txt tss/360 360/67 were sold to lots of univ. ... but TSS/360 never quite became product ... and so many systems ran as 360/65 as os/360 for most of the time. as undergraduate in the 60s, there was work on both tss/360 and cp67 on the weekends. we did one fortran edit, compile, and execute simulated user benchmark. tss/360 running the script with four simulated users got worse throughput and interactive response than cp67/cms with 35 simulated users (on the same hardware). the Future System project in the early 70s was going to also be single-level-store based ... from tss/360, multics, etc. recent reference to future system: http://www.garlic.com/~lynn/2013h.html#44 Why does IBM keep saying things like this: at the same time I was at the science center and did paged-mapped filesystem for cms (somewhat in competition with multics on the floor above). Having observed a lot of the tss/360 problems ... I did an implementation that avoided many of the problems (and would periodically ridicule the FS ... claiming what I already had running was better). it never made it as part of release product ... in part because of the bad rep that single-level-store got from the FS effort ... even though I could show 3times the throughput/efficiency compared to standard CMS filesystem (both CDF EDF) for moderate filesystem workload. some past posts http://www.garlic.com/~lynn/submain.html#mmap part of it was to have same exact shared segments at different virtual addresses in different virtual address spaces ... which also wasn't released. A small piece of the shared segment (w/o filesystem stsuff and supporting concurrent shared images at different virtual addresses) ... was released as DCSS in vm370 release 3. some old email from the period http://www.garlic.com/~lynn/2006v.html#email731212 http://www.garlic.com/~lynn/2006w.html#email750102 http://www.garlic.com/~lynn/2006w.html#email750430 one of the hardest problems was that CMS borrowed a lot of stuff from os/360 and there was enormous problem with os/360 relocatable adcons (which are swizzled to absolute address after being loaded for execution). I needed *real* relocatable adcons ... both at load time as well at execution time. This was one thing supported by tss/360. some past posts discussing constant battle that i had with os/360 relocatable adcons http://www.garlic.com/~lynn/submain.html#adcon note that folklore is that after FS failure, some of the people retreated to Rochester and did single-level-store for S/38 ... note however, the S/38 wasn't into performance throughput ... so the single-level-store performance issues weren't an issue. -- virtualization experience starting Jan1968, online at home since Mar1970 -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Storage paradigm [was: RE: Data volumes]
l...@garlic.com (Anne Lynn Wheeler) writes: it never made it as part of release product ... in part because of the bad rep that single-level-store got from the FS effort ... even though I could show 3times the throughput/efficiency compared to standard CMS filesystem (both CDF EDF) for moderate filesystem workload. some past posts http://www.garlic.com/~lynn/submain.html#mmap re: http://www.garlic.com/~lynn/2013h.html#44 Why does IBM keep saying things like this http://www.garlic.com/~lynn/2013h.html#45 Storage paradigm [was: RE: Data volumes] note that there was two parts of the significant throughput increase going to paged-mapped filesystem ... one is that it is higher level abstraction that allows significant amount of optimization on service the request to be done under the covers. the other is there is a real paradigm mismatch between channel program paradigm and virtual memory ... requiring significant pathlengths to stitch together the mismatch paradigm match/mismatch continues on down ... there is almost direct paradigm match with page mapped filesystem at the top, down through virtual memory operation to fixed-block architecture disk structure; and when there is mismatch requires a lot of extra resources and overhead ... like effort to map CKD to FBA (aka there hasn't been any real CKD disks manufactured for decades). Virtual machines has to scan each channel program, making a copy and replacing virtual addresses for real. CMS standard filesystem used CCWs for i/o ... it was originally developed on the same 360/40 (running stand-alone) as was being used to do CP40 (with added virtual memory hardware). CMS continued to be able to run on bare machine all during cp67 ... but there was an artificial cripple put in in the morph from vm370/cms. In the transition from MVT to OS/VS2 (aka virtual memory), the same problem showed up. The original implementation involved putting a little bit of code to create 16mbyte virtual address space for MVT, but the major effort was hacking CCWTRANS (from CP67) into the side of EXCP processing (EXCP had the same problem with access methods creating channel programs in the application virtual address space ... as CP67 did with virtual machine channel programs). Old reference by somebody in POK that was in the middle of the transition to OS/VS2 and virtual memory ... includes reference that OS/VS2 release 2 (MVS) was on glide path to OS/VS2 release 3 (FS) http://www.garlic.com/~lynn/2011d.html#73 Multiple Virtual Memory above also mentions that one of the people responsible for HASP having a group doing a TSS-style implementation of OS/360 (that had something like a page-mapped filesystem) One of the big differences (cp67/cms and os/vs2) was that the rest of CP67 I/O processing was as little as 5% of the corresponding OS/VS2 pathlength ... so the additional pathlength for doing channel program duplication with real addresses was less noticeable in OS/VS2 (both SVS and MVS). In fact, one of the big motivations for the SSCH and other changes to I/O for 370-xa was to get some portion of the enormous I/O pathlength moved out of MVS so it could be rewritten (as well as being moved to separate dedicated processors). Starting in the late 70s, I got to play disk engineer in the disk engineering labs and rewrite the I/O supervisor to be bullet proof and never fail ... so they could do concurrent on-demand development testing in operating system environment (they had tried MVS, but found MVS had 15min MTBF in that environment with just a single testcell ... requiring manual re-ipl). However, I also attempted to do further pathlenth reducution (while still supporting never fail) to demonstrate 370 I/O coming as close as possible to 370-xa with separate dedicated processor. past posts mentioning getting to play disk engineer http://www.garlic.com/~lynn/subtopic.html#disk for other topic drift ... FE had 3380 i/o error injection regression tests and even after 3380 was introduced, MVS was failing (requiring re-ipl) for all tests ... and in 2/3rds of the tests, there was no indication what had precipitated the failure. old email: http://www.garlic.com/~lynn/2007.html#email801015 -- virtualization experience starting Jan1968, online at home since Mar1970 -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Data volumes
On 6/9/2013 10:07 AM, John Gilmore wrote: The numbers 9 and 27 (for the venturesome) still prevail in many shops. Much larger values, never available as physical, spinning DASD, appear to make some people uneasy. They get used, reluctantly at least on the first occasion, only when some compelling special argument is made for them. In querying people about such decisions I have been told things like The smaller volumes are more reliable or I wasn't sure the big ones would work. In our case management wouldn't buy PAV, so we don't use really large volumes to avoid contention. -- Richard -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Data volumes
Richard, I don't of course know the details of your situation, but in each of the 6 six rather different shops where I have looked at this issue in detail it has turned out that PAV was the economic choice, decisively so. If an occasion for revisiting it quantitatively ever arises, you should try to get your management to have another look at this decision in changed circumstances. John Gilmore, Ashland, MA 01721 - USA -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Storage paradigm [was: RE: Data volumes]
Rant Like a few others on this list, I have often gritted my teeth at the necessity to estimate disk storage quantities that vary widely over time in a fixed manner (i.e., SPACE in JCL) when the true need is just to match output volume to input volume each day. Why is it that IBM (and organizations that use their mainframe systems) so vigorously resist a conversion off of the ECKD standard? (Yes, I know it's all about conversion cost, but in the larger picture that is a red herring.) Not that I'm likely to see such a transition in my lifetime, but in this dawning time of soi-disant big data, perhaps it is past time to change the storage paradigm entirely, not from ECKD to FBA but to transition instead to something like the Multics model where every object in the system (whether in memory or on external storage, whether data or program) has an address, and all addresses are unique. Let the storage subsystem decide how to optimally position and aggregate the various parts of objects, and how to organize them for best performance. Such decisions should not require human guesstimate input to be optimal, or nearly so. Characteristics of application access are far more critical specifications than mere size. The ability to specify just the desired application access characteristics (random, sequential, growing, shrinking, response-time-critical, etc.) should be necessary and sufficient. EAV or not EAV, guaranteed space or not, candidate volumes, striped or not striped, compressed or not compressed - all of that baggage is clearly non-optimal for getting the job done in a timely manner. Why should allocating a simple sequential file require a team of Storage Administration experts to accomplish effectively? /Rant Peter -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Ed Jaffe Sent: Sunday, June 09, 2013 10:47 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: Data volumes On 6/9/2013 7:12 AM, Scott Ford wrote: We need bigger dasd ...ouch The largest 3390 volumes in our tiny shop hold 3,940,020 tracks or 262,668 cylinders. That is the maximum size supported by the back-level DASD we are running. Newer DASD hardware can support volumes up to 1TB in size. I assume nearly all zEC12 and z196 customers are capable of exploiting these large sizes. But, do they? I spent three years dealing with, and eventually helping IBM to solve (via OA40210 - HIPER, DATALOSS), a serious EAV bug that should have been seen in most every shop in the world that uses the DFSMSdss CONSOLIDATE function (with or without DEFRAG). The experience was a real eye-opener for me and I concluded that almost nobody is using EAV! Why not? Personally, I would find it embarrassing if the Corsair thumb drive in my pocket held more data than our largest mainframe volumes. But, that's just me... -- This message and any attachments are intended only for the use of the addressee and may contain information that is privileged and confidential. If the reader of the message is not the intended recipient or an authorized representative of the intended recipient, you are hereby notified that any dissemination of this communication is strictly prohibited. If you have received this communication in error, please notify us immediately by e-mail and delete the message and any attachments from your system. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Storage paradigm [was: RE: Data volumes]
In general, I agree. But I will say that I need something to limit run-away usage of disk space. Why? Because we have had programmers who didn't want to be bother either. So they put out a report to SPOOL. And then their program went into a loop; writing the same message over and over. This exhausted the SPOOL space. Which caused a production outage on a weekend (we run dark on the weekends). I can easily envision the same thing happening if DASD ever went to an all you can eat. Of course, with SMS control, it is easier to segregate the data into pools. We currently try to do a type of all you can reasonably eat by having our data classes have a dynamic volume count of 59. And we have a semi-standard (read: recommendation) that files of an unknown size be allocated CYL,(500,100) . Oh, and we use space release in the data class to get rid of excess allocation. If the data set cannot abide space release for some reason, there is an exempt data class for the programmers to use. On Mon, Jun 10, 2013 at 10:38 AM, Farley, Peter x23353 peter.far...@broadridge.com wrote: Rant Like a few others on this list, I have often gritted my teeth at the necessity to estimate disk storage quantities that vary widely over time in a fixed manner (i.e., SPACE in JCL) when the true need is just to match output volume to input volume each day. Why is it that IBM (and organizations that use their mainframe systems) so vigorously resist a conversion off of the ECKD standard? (Yes, I know it's all about conversion cost, but in the larger picture that is a red herring.) Not that I'm likely to see such a transition in my lifetime, but in this dawning time of soi-disant big data, perhaps it is past time to change the storage paradigm entirely, not from ECKD to FBA but to transition instead to something like the Multics model where every object in the system (whether in memory or on external storage, whether data or program) has an address, and all addresses are unique. Let the storage subsystem decide how to optimally position and aggregate the various parts of objects, and how to organize them for best performance. Such decisions should not require human guesstimate input to be optimal, or nearly so. Characteristics of application access are far more critical specifications than mere size. The ability to specify just the desired application access characteristics (random, sequential, growing, shrinking, response-time-critical, etc.) should be necessary and sufficient. EAV or not EAV, guaranteed space or not, candidate volumes, striped or not striped, compressed or not compressed - all of that baggage is clearly non-optimal for getting the job done in a timely manner. Why should allocating a simple sequential file require a team of Storage Administration experts to accomplish effectively? /Rant Peter -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Ed Jaffe Sent: Sunday, June 09, 2013 10:47 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: Data volumes On 6/9/2013 7:12 AM, Scott Ford wrote: We need bigger dasd ...ouch The largest 3390 volumes in our tiny shop hold 3,940,020 tracks or 262,668 cylinders. That is the maximum size supported by the back-level DASD we are running. Newer DASD hardware can support volumes up to 1TB in size. I assume nearly all zEC12 and z196 customers are capable of exploiting these large sizes. But, do they? I spent three years dealing with, and eventually helping IBM to solve (via OA40210 - HIPER, DATALOSS), a serious EAV bug that should have been seen in most every shop in the world that uses the DFSMSdss CONSOLIDATE function (with or without DEFRAG). The experience was a real eye-opener for me and I concluded that almost nobody is using EAV! Why not? Personally, I would find it embarrassing if the Corsair thumb drive in my pocket held more data than our largest mainframe volumes. But, that's just me... -- This message and any attachments are intended only for the use of the addressee and may contain information that is privileged and confidential. If the reader of the message is not the intended recipient or an authorized representative of the intended recipient, you are hereby notified that any dissemination of this communication is strictly prohibited. If you have received this communication in error, please notify us immediately by e-mail and delete the message and any attachments from your system. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- This is a test of the Emergency Broadcast System. If this had been an actual emergency, do you really think we'd stick around to tell you? Maranatha! John McKown -- For IBM-MAIN subscribe
Re: Storage paradigm [was: RE: Data volumes]
On Mon, 10 Jun 2013 11:38:08 -0400, Farley, Peter x23353 peter.far...@broadridge.com wrote: Rant Like a few others on this list, I have often gritted my teeth at the necessity to estimate disk storage quantities that vary widely over time in a fixed manner (i.e., SPACE in JCL) when the true need is just to match output volume to input volume each day. Why is it that IBM (and organizations that use their mainframe systems) so vigorously resist a conversion off of the ECKD standard? (Yes, I know it's all about conversion cost, but in the larger picture that is a red herring.) Not that I'm likely to see such a transition in my lifetime, but in this dawning time of soi-disant big data, perhaps it is past time to change the storage paradigm entirely, not from ECKD to FBA but to transition instead to something like the Multics model where every object in the system (whether in memory or on external storage, whether data or program) has an address, and all addresses are unique. Let the storage subsystem decide how to optimally position and aggregate the various parts of objects, and how to organize them for best performance. Such decisions should not require human guesstimate input to be optimal, or nearly so. Characteristics of application access are far more critical specifications than mere size. The ability to specify just the desired application access characteristics (random, sequential, growing, shrinking, response-time-critical, etc.) should be necessary and sufficient. EAV or not EAV, guaranteed space or not, candidate volumes, striped or not striped, compressed or not compressed - all of that baggage is clearly non-optimal for getting the job done in a timely manner. Why should allocating a simple sequential file require a team of Storage Administration experts to accomplish effectively? /Rant Peter Oh, then you want to move to IBM System i... :-) Seriously, System i, (formerly known by many different names), addresses everything in storage/memory and on disk and has been 64-bit since the mid 1990s, (and it was 48-bit before that). There is however, no need for system programmers on the i, (really IBM, you can't come up with better names for hardware/software than 1 character? And i, is this an Apple box?) -- Dale R. Smith -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Storage paradigm [was: RE: Data volumes]
I am not a LUW person, other than I use a windows machine for simple things, so I am curious how external storage is allocated and controlled in that environment. I think we have all heard the complaints about the short-comings of MVS in this area, but what would be a realistic solution? I would imagine the people at IBM have spent a little time on this, and if it was easy would have started transitioning us from ECKD to 'the new way' a long time ago. The idea of a 'run-away' program is what is the hang-up. Chris Blaicher Principal Software Engineer, Software Development Syncsort Incorporated 50 Tice Boulevard, Woodcliff Lake, NJ 07677 P: 201-930-8260 | M: 512-627-3803 E: cblaic...@syncsort.com -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Farley, Peter x23353 Sent: Monday, June 10, 2013 10:38 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Storage paradigm [was: RE: Data volumes] Rant Like a few others on this list, I have often gritted my teeth at the necessity to estimate disk storage quantities that vary widely over time in a fixed manner (i.e., SPACE in JCL) when the true need is just to match output volume to input volume each day. Why is it that IBM (and organizations that use their mainframe systems) so vigorously resist a conversion off of the ECKD standard? (Yes, I know it's all about conversion cost, but in the larger picture that is a red herring.) Not that I'm likely to see such a transition in my lifetime, but in this dawning time of soi-disant big data, perhaps it is past time to change the storage paradigm entirely, not from ECKD to FBA but to transition instead to something like the Multics model where every object in the system (whether in memory or on external storage, whether data or program) has an address, and all addresses are unique. Let the storage subsystem decide how to optimally position and aggregate the various parts of objects, and how to organize them for best performance. Such decisions should not require human guesstimate input to be optimal, or nearly so. Characteristics of application access are far more critical specifications than mere size. The ability to specify just the desired application access characteristics (random, sequential, growing, shrinking, response-time-critical, etc.) should be necessary and sufficient. EAV or not EAV, guaranteed space or not, candidate volumes, striped or not striped, compressed or not compressed - all of that baggage is clearly non-optimal for getting the job done in a timely manner. Why should allocating a simple sequential file require a team of Storage Administration experts to accomplish effectively? /Rant Peter -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Ed Jaffe Sent: Sunday, June 09, 2013 10:47 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: Data volumes On 6/9/2013 7:12 AM, Scott Ford wrote: We need bigger dasd ...ouch The largest 3390 volumes in our tiny shop hold 3,940,020 tracks or 262,668 cylinders. That is the maximum size supported by the back-level DASD we are running. Newer DASD hardware can support volumes up to 1TB in size. I assume nearly all zEC12 and z196 customers are capable of exploiting these large sizes. But, do they? I spent three years dealing with, and eventually helping IBM to solve (via OA40210 - HIPER, DATALOSS), a serious EAV bug that should have been seen in most every shop in the world that uses the DFSMSdss CONSOLIDATE function (with or without DEFRAG). The experience was a real eye-opener for me and I concluded that almost nobody is using EAV! Why not? Personally, I would find it embarrassing if the Corsair thumb drive in my pocket held more data than our largest mainframe volumes. But, that's just me... -- This message and any attachments are intended only for the use of the addressee and may contain information that is privileged and confidential. If the reader of the message is not the intended recipient or an authorized representative of the intended recipient, you are hereby notified that any dissemination of this communication is strictly prohibited. If you have received this communication in error, please notify us immediately by e-mail and delete the message and any attachments from your system. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN ATTENTION: - The information contained in this message (including any files transmitted with this message) may contain proprietary, trade secret or other confidential and/or legally privileged information. Any pricing information contained in this message or in any files transmitted with this message is always confidential and cannot be shared with any third parties without prior written
Re: Storage paradigm [was: RE: Data volumes]
LUW works similar to z/OS UNIX file systems. I.e. there is a file system which is formatted using some utility (mkfs in the Linux/UNIX world, format in Windows). This sets up all the internals. In today's LUW, it is usually possible for a single file to be as big as the file system upon which it resides. But no bigger. There is nothing like a multi file system file (which would vaguely like a multivolume data set). I'm not too Windows literate any more, but it used to be that a file system had to reside on a single disk (or in a partition of that disk). Linux/UNIX used to be that way too. But Linux/UNIX now implements something called LVM (Logical Volume Manager). In short, LVM can stitch together a number of physical disk volumes (called a PV for Physical Volume), or partitions, and the subdivide that aggregate space into one or more Logical Volumes (LV). A Logical Volume can be created in many ways, such as using software RAID, or striping. The admin then formats a file system on the Logical Volume. Even after creating the PV, the storage admin can add another disk into a PV (like adding a volume to a storage group in SMS). The storage admin can then use the new space for another LV or to extend an existing LV. Depending on the file system formatted on the LV, it might even be possible to tell the file system to start using the newly added space. Most of the current Linux file systems can at least be extended when they are unmounted (not actively used). So, like a zFS file system on z/OS, using something like EXT4 (or BTRFS) and LVM, it is possible to dynamically expand the size of a file system. I don't know for certain, but I doubt that Windows can do this kind of dynamic expansion at all. To control allocation, Linux and UNIX can use disk quotas. I don't know much about that since I don't use it on my personal systems. On Mon, Jun 10, 2013 at 11:15 AM, Blaicher, Christopher Y. cblaic...@syncsort.com wrote: I am not a LUW person, other than I use a windows machine for simple things, so I am curious how external storage is allocated and controlled in that environment. I think we have all heard the complaints about the short-comings of MVS in this area, but what would be a realistic solution? I would imagine the people at IBM have spent a little time on this, and if it was easy would have started transitioning us from ECKD to 'the new way' a long time ago. The idea of a 'run-away' program is what is the hang-up. Chris Blaicher Principal Software Engineer, Software Development Syncsort Incorporated 50 Tice Boulevard, Woodcliff Lake, NJ 07677 P: 201-930-8260 | M: 512-627-3803 E: cblaic...@syncsort.com -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Farley, Peter x23353 Sent: Monday, June 10, 2013 10:38 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Storage paradigm [was: RE: Data volumes] Rant Like a few others on this list, I have often gritted my teeth at the necessity to estimate disk storage quantities that vary widely over time in a fixed manner (i.e., SPACE in JCL) when the true need is just to match output volume to input volume each day. Why is it that IBM (and organizations that use their mainframe systems) so vigorously resist a conversion off of the ECKD standard? (Yes, I know it's all about conversion cost, but in the larger picture that is a red herring.) Not that I'm likely to see such a transition in my lifetime, but in this dawning time of soi-disant big data, perhaps it is past time to change the storage paradigm entirely, not from ECKD to FBA but to transition instead to something like the Multics model where every object in the system (whether in memory or on external storage, whether data or program) has an address, and all addresses are unique. Let the storage subsystem decide how to optimally position and aggregate the various parts of objects, and how to organize them for best performance. Such decisions should not require human guesstimate input to be optimal, or nearly so. Characteristics of application access are far more critical specifications than mere size. The ability to specify just the desired application access characteristics (random, sequential, growing, shrinking, response-time-critical, etc.) should be necessary and sufficient. EAV or not EAV, guaranteed space or not, candidate volumes, striped or not striped, compressed or not compressed - all of that baggage is clearly non-optimal for getting the job done in a timely manner. Why should allocating a simple sequential file require a team of Storage Administration experts to accomplish effectively? /Rant Peter -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Ed Jaffe Sent: Sunday, June 09, 2013 10:47 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: Data volumes On 6/9/2013 7:12 AM, Scott Ford wrote: We need bigger dasd ...ouch
Re: Storage paradigm [was: RE: Data volumes]
For Windows Capabilities, I suggest reading about Dynamic Disks and Dynamic Volumes on MSDN: http://msdn.microsoft.com/en-us/library/windows/desktop/aa363785(v=vs.85).aspx John -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Storage paradigm [was: RE: Data volumes]
On 6/10/2013 12:15 PM, Blaicher, Christopher Y. wrote: I am not a LUW person, other than I use a windows machine for simple things, so I am curious how external storage is allocated and controlled in that environment. I think we have all heard the complaints about the short-comings of MVS in this area, but what would be a realistic solution? Technically the easiest to implement would be adding a new device type, thus keeping (E)CKD completely distinct from FBA. The new type could be supported by VSAM/AMS only (and JCL, SVC 99, etc.) without impacting other programs. (I would hope there are no programs out there using TM UCBTBYT3 rather than CLI?) I would imagine the people at IBM have spent a little time on this, and if it was easy would have started transitioning us from ECKD to 'the new way' a long time ago. The idea of a 'run-away' program is what is the hang-up. I cannot see IBM spend any effort on this unless one of the Fortune 10 companies requires it. Even then I would expect them to stick with the pre-allocated space paradigm rather than transitioning. As to run-away programs, they should be thoroughly checked on a test system before going into production; a run-away in production should be so rare as to be immaterial. Gerhard Postpischil Bradford, Vermont -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Storage paradigm [was: RE: Data volumes]
On 6/10/2013 11:38 AM, Farley, Peter x23353 wrote: Rant Like a few others on this list, I have often gritted my teeth at the necessity to estimate disk storage quantities that vary widely over time in a fixed manner (i.e., SPACE in JCL) when the true need is just to match output volume to input volume each day. If it's that predictable, it's trivial to write code to produce an estimated output volume from input, and tailor and submit the appropriate JCL. So that's a non-issue. EAV or not EAV, guaranteed space or not, candidate volumes, striped or not striped, compressed or not compressed - all of that baggage is clearly non-optimal for getting the job done in a timely manner. Why should allocating a simple sequential file require a team of Storage Administration experts to accomplish effectively? /Rant There is no theoretical solution. On any system running jobs, it is possible for one job to monopolize available space, requiring other jobs to wait forever or be terminated. Even on a single job system that job may exhaust space. Requiring a space specification may be a PITA, but it guarantees that a started job will finish (subject to other constraints). And the SA experts, especially for sequential files, can be avoided with simple estimator programs. This seems to be more of a religious war than a practical discussion. Gerhard Postpischil Bradford, Vermont -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Storage paradigm [was: RE: Data volumes]
Download GnuPartEd, Burn it to CD-ROM, Boot from it, resize as needed. On Mon, Jun 10, 2013 at 11:56 AM, Roberts, John J jrobe...@dhs.state.ia.us wrote: For Windows Capabilities, I suggest reading about Dynamic Disks and Dynamic Volumes on MSDN: http://msdn.microsoft.com/en-us/library/windows/desktop/aa363785(v=vs.85).aspx John -- 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: Storage paradigm [was: RE: Data volumes]
As to the religious aspect, I did try to signal the less-than-practical nature of my note with the Rant and /Rant tags. To your point about tailoring and dynamically submitting JCL, it really is an issue. In a typical large z/OS shop today, dynamically tailoring and submitting JCL is only permitted for test environments and users. Production JCL is frozen and controlled and submitted only by the scheduler software, and there is no political possibility to dynamically adjust the parameters even if it is technically feasible. There *are* non-theoretical solutions to runaway file output. The *ix system model of using disk quotas per user makes it entirely possible to imagine z/OS application users with reasonable disk quotas specific to the application (i.e., not by job but by suite of jobs). Not the best solution? Maybe not, but ISTM to be better than having to predict what each and every process (i.e., job and file) output volume will be. And there may well be other process models out there different from anything I know or imagine. I don't claim to have an exclusive lock on ideas to replace what we have to deal with. Peter -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf Of Gerhard Postpischil Sent: Monday, June 10, 2013 1:32 PM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: Storage paradigm [was: RE: Data volumes] On 6/10/2013 11:38 AM, Farley, Peter x23353 wrote: Rant Like a few others on this list, I have often gritted my teeth at the necessity to estimate disk storage quantities that vary widely over time in a fixed manner (i.e., SPACE in JCL) when the true need is just to match output volume to input volume each day. If it's that predictable, it's trivial to write code to produce an estimated output volume from input, and tailor and submit the appropriate JCL. So that's a non-issue. EAV or not EAV, guaranteed space or not, candidate volumes, striped or not striped, compressed or not compressed - all of that baggage is clearly non-optimal for getting the job done in a timely manner. Why should allocating a simple sequential file require a team of Storage Administration experts to accomplish effectively? /Rant There is no theoretical solution. On any system running jobs, it is possible for one job to monopolize available space, requiring other jobs to wait forever or be terminated. Even on a single job system that job may exhaust space. Requiring a space specification may be a PITA, but it guarantees that a started job will finish (subject to other constraints). And the SA experts, especially for sequential files, can be avoided with simple estimator programs. This seems to be more of a religious war than a practical discussion. Gerhard Postpischil Bradford, Vermont -- This message and any attachments are intended only for the use of the addressee and may contain information that is privileged and confidential. If the reader of the message is not the intended recipient or an authorized representative of the intended recipient, you are hereby notified that any dissemination of this communication is strictly prohibited. If you have received this communication in error, please notify us immediately by e-mail and delete the message and any attachments from your system. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Storage paradigm [was: RE: Data volumes]
Too true. And, around here, our QA people appear to be glitz checkers instead of function and reliability checkers. They have more people than any other group and do less testing on the mainframe. They seem to check mainly for ease of use. That is, can a totally numb skull still use this? On Mon, Jun 10, 2013 at 1:57 PM, Ted MacNEIL eamacn...@yahoo.ca wrote: As to run-away programs, they should be thoroughly checked on a test system before going into production; a run-away in production should be so rare as to be immaterial. What planet are you from? Programmers seem able to test everything except that one condition that will break in Production - Ted MacNEIL eamacn...@yahoo.ca Twitter: @TedMacNEIL -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- This is a test of the Emergency Broadcast System. If this had been an actual emergency, do you really think we'd stick around to tell you? Maranatha! John McKown -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: [SPAM] Re: Storage paradigm [was: RE: Data volumes]
On 6/10/2013 2:46 PM, Farley, Peter x23353 wrote: To your point about tailoring and dynamically submitting JCL, it really is an issue. In a typical large z/OS shop today, dynamically tailoring and submitting JCL is only permitted for test environments and users. Production JCL is frozen and controlled and submitted only by the scheduler software, and there is no political possibility to dynamically adjust the parameters even if it is technically feasible. The scheduler can be set up to submit the tailoring job just as easily as the job to be tailored. And a few critical production abends should take care of the political aspect. There *are* non-theoretical solutions to runaway file output. The *ix system model of using disk quotas per user makes it entirely possible to imagine z/OS application users with reasonable disk quotas specific to the application (i.e., not by job but by suite of jobs). Not the best solution? Maybe not, but ISTM to be better than having to predict what each and every process (i.e., job and file) output volume will be. I've worked for service bureaus that established just such quotas. My objections stand, as both the single job and a suite of jobs can still fail; the difference is the number of jobs/users impacted, not the principle. And there may well be other process models out there different from anything I know or imagine. I don't claim to have an exclusive lock on ideas to replace what we have to deal with Ditto. Gerhard Postpischil Bradford, Vermont -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Storage paradigm [was: RE: Data volumes]
On 6/10/2013 2:57 PM, Ted MacNEIL wrote: What planet are you from? Sol 3 Programmers seem able to test everything except that one condition that will break in Production Perhaps you are keeping bad company. While humans are not perfect, there are methods to improve code reliability. Gerhard Postpischil Bradford, Vermont -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Data volumes
In an article in this morning's New York Times that is chiefly about the new NSA complex in Utah I found: begin extract Because of smartphones, tablets, social media sites, e-mail and other forms of digital communications, the world creates 2.5 quintillion bytes of new data daily, according to I.B.M. /end extract Times copy editors apparently judged that 'quintillion' would be more accessible to readers, but 1 q John Gilmore, Ashland, MA 01721 - USA -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Data volumes
continuing . . . 1 quintillion = 1,000,000,000,000,000,000 = 1 exabyte -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Data volumes
We need bigger dasd ...ouch Scott ford www.identityforge.com from my IPAD 'Infinite wisdom through infinite means' On Jun 9, 2013, at 6:45 AM, John Gilmore jwgli...@gmail.com wrote: continuing . . . 1 quintillion = 1,000,000,000,000,000,000 = 1 exabyte -- 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: Data volumes
On 6/9/2013 7:12 AM, Scott Ford wrote: We need bigger dasd ...ouch The largest 3390 volumes in our tiny shop hold 3,940,020 tracks or 262,668 cylinders. That is the maximum size supported by the back-level DASD we are running. Newer DASD hardware can support volumes up to 1TB in size. I assume nearly all zEC12 and z196 customers are capable of exploiting these large sizes. But, do they? I spent three years dealing with, and eventually helping IBM to solve (via OA40210 - HIPER, DATALOSS), a serious EAV bug that should have been seen in most every shop in the world that uses the DFSMSdss CONSOLIDATE function (with or without DEFRAG). The experience was a real eye-opener for me and I concluded that almost nobody is using EAV! Why not? Personally, I would find it embarrassing if the Corsair thumb drive in my pocket held more data than our largest mainframe volumes. But, that's just me... -- Edward E Jaffe Phoenix Software International, Inc 831 Parkview Drive North El Segundo, CA 90245 http://www.phoenixsoftware.com/ -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Data volumes
The numbers 9 and 27 (for the venturesome) still prevail in many shops. Much larger values, never available as physical, spinning DASD, appear to make some people uneasy. They get used, reluctantly at least on the first occasion, only when some compelling special argument is made for them. In querying people about such decisions I have been told things like The smaller volumes are more reliable or I wasn't sure the big ones would work. John Gilmore, Ashland, MA 01721 - USA -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: Data volumes
On Sun, 9 Jun 2013 11:07:39 -0400, John Gilmore wrote: The numbers 9 and 27 (for the venturesome) still prevail in many shops. Much larger values, never available as physical, spinning DASD, appear to make some people uneasy. Too true. Being somewhat of an itinerant, I see this a bit. Mind you, the (best) functionality tends to be on only the latest kit IBM want to flog, or for the other vendors, maybe as an added cost feature. Even getting (hiper)PAV has been a struggle at some sites. Shane ... -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN