Re: Acceptable paging
At least with z196 I think the cost of real memory in the processor is low enough that you should be able to make a case to purchase more to reduce or eliminate paging if it is a performance issue. We manage to 0% residency on local page data sets instead manage real storage available budget by AFQA size. The instructions to do that paging I/O take elapsed time, execute on general purpose processors, and the cycle time out to cache in a storage processor is an order of magnitude greater than reference to an old but paged in memory frame. Memory rich environments don't have to be all that rich with real memory on z196 now roughly 15% of the cost of what it was when IBM announced the mainframe charter in 2003. Best Regards, Sam Knutson, GEICO System z Team Leader mailto:sknut...@geico.com (office) 301.986.3574 (cell) 301.996.1318 Think big, act bold, start simple, grow fast... -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of Vernooij, CP - SPLXM Sent: Monday, February 13, 2012 2:37 AM To: IBM-MAIN@bama.ua.edu Subject: Re: Acceptable paging Ron, At the moment I only have ESSs, but we are going to new storage soon. We expect everything to improve. Kees. Ron Hawkins ronjhawk...@sbcglobal.net wrote in message news:00ad01cce88a$d99abf30$8cd03d90$@net... Kees, Have you thought of putting your page datasets into Permacache (EMC) or DCR (HDS). This would give you true SSD performance right on the channel, which is better than the SAS HDD emulation variety. Ron This email/fax message is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution of this email/fax is prohibited. If you are not the intended recipient, please destroy all paper and electronic copies of the original message. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Acceptable paging
Ron, At the moment I only have ESSs, but we are going to new storage soon. We expect everything to improve. Kees. Ron Hawkins ronjhawk...@sbcglobal.net wrote in message news:00ad01cce88a$d99abf30$8cd03d90$@net... Kees, Have you thought of putting your page datasets into Permacache (EMC) or DCR (HDS). This would give you true SSD performance right on the channel, which is better than the SAS HDD emulation variety. Ron Dead, until you stop the Execution Groups. For some reason this storage is then needed and paged in again, which turned up in delays during shutdown and brought 1 or 2 bad performing hotspots in my paging configuration above the table (or is this typically Dutch). Kees. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.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...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Acceptable paging
Kees, Have you thought of putting your page datasets into Permacache (EMC) or DCR (HDS). This would give you true SSD performance right on the channel, which is better than the SAS HDD emulation variety. Ron Dead, until you stop the Execution Groups. For some reason this storage is then needed and paged in again, which turned up in delays during shutdown and brought 1 or 2 bad performing hotspots in my paging configuration above the table (or is this typically Dutch). Kees. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Acceptable paging
Barbara Nitz nitz-...@gmx.net wrote in message news:863014571846.wa.nitzibmgmx@bama.ua.edu... Maybe you can explain what MQ and broker issues you have and why you think this is related to paging and/or the page configuration. I agree with Kees. What exactly is your 'issue' with 'the broker'? (Assuming that MQS is just the innocent bystander here.) I know the broker uses a lot of virtual storage, which indeed requires additional paging space, so the first check you can do is check how full your paging configuration is. 'The broker' uses a lot of shared storage (SHRLIBRGNSIZE), and it has a bad habit of polling for work (i.e., in all those numbered address spaces' timers pop at the same time), generating artificial 'loads' that WLM cannot keep up with. In our production environment 'the brokers' generally have a PI in the double digits (yes, double digit PI, sometimes up to triple digits - a bad clicker application NOT designed for resource sharing, i.e. z/OS). From what I understand, in the newest release they also 'require' a lot of 64bit storage, possibly due to some sort of websphere application server hidden in there (but I may be wrong on this). I also have the sneaking suspicion (which I am currently also investigating) that 'the broker' goes out and gets itself a lot of (virtual) storage, touches each and every of those pages and then never uses them. Which means they become slots out on AUX, or rather, they become dead weight. Dead, until you stop the Execution Groups. For some reason this storage is then needed and paged in again, which turned up in delays during shutdown and brought 1 or 2 bad performing hotspots in my paging configuration above the table (or is this typically Dutch). Kees. What's a few more Gig of storage to a clicker, anyway? If your issue is 'only' paging, then do a D ASM and add local page space until % usage is way below 30%. Make sure that you have your PAGTOTL in ieasys sufficiently high, as that can only be changed via IPL. Barbara -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.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...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Acceptable paging
ADDRESS SPACE! snip From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of Shmuel Metz (Seymour J.) Sent: Wednesday, February 08, 2012 8:44 PM To: IBM-MAIN@bama.ua.edu Subject: Re: Acceptable paging In 1910aea19cd2554fb59403184ebe4381044b40a...@mmoexchmbs01.jhacorp.com, on 02/08/2012 at 04:49 PM, Hal Merritt hmerr...@jackhenry.com said: A demand page in event can cause a full stop in the requesting address space Address space or only task? /snip -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Acceptable paging
Can someone please help me determine what the 'acceptable paging' should be on a system? We are having issues with MQ and broker. Thanks. Could you describe your environment a little? CPU(s) (z10, z9 , etc) Amount of dasd space allocated to Page Datasets? Number of Page Datasets Real Memory available Number of MQ and Brokers? Connection types (IMS, CICS, Distributed systems) Many elements can play into tuning a system. This will help in answering the question. Now, when you ask about acceptable paging in relationship to MQ, is anything else in play? CICS, IMS, DB2, heavy batch, etc. or is MQ the only thing running on that system? Or perhaps I misunderstood your acceptable paging in relationship to MQ. Thanks Lizette -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Acceptable paging
We have a z/10 and this test lpar had 11G of real memory. IBM is loaning us 16G of real memory so now we have 27G of real memory. We had 3 mod-9 page datasets and after message broker started we had to add 4 additional mod-9 page datasets. We do have IMS, CICS, and DB2 running on this system. Our tuning guy retired last year and none of us left understand how this is supposed to work. Thanks. -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of Lizette Koehler Sent: Wednesday, February 08, 2012 9:11 AM To: IBM-MAIN@bama.ua.edu Subject: Re: Acceptable paging Can someone please help me determine what the 'acceptable paging' should be on a system? We are having issues with MQ and broker. Thanks. Could you describe your environment a little? CPU(s) (z10, z9 , etc) Amount of dasd space allocated to Page Datasets? Number of Page Datasets Real Memory available Number of MQ and Brokers? Connection types (IMS, CICS, Distributed systems) Many elements can play into tuning a system. This will help in answering the question. Now, when you ask about acceptable paging in relationship to MQ, is anything else in play? CICS, IMS, DB2, heavy batch, etc. or is MQ the only thing running on that system? Or perhaps I misunderstood your acceptable paging in relationship to MQ. Thanks Lizette -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN E-mail correspondence to and from this address may be subject to the North Carolina Public Records Law and may be disclosed to third parties by an authorized state official. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Acceptable paging
On Wed, 8 Feb 2012 14:01:48 +, Lopez, Sharon sharon.lo...@nc.gov wrote: Can someone please help me determine what the 'acceptable paging' should be on a system? We are having issues with MQ and broker. Thanks. Any amount that still keeps your end user response time and job turnaround times within your SLAs. If you have no SLAs then whatever keeps your phone from ringing. :-) That being said, as inexpensive as memory is these days and the benefits of reduced I/O by taking advantage of features within the OS that utilize more memory, I think most shops try to keep their demand paging close to zero for production LPARs. One of the LPARs I support is a WAS development LPAR with 183 control and servant regions and even though it has 80G of real storage the average demand paging rate is probably between 10-30 and at times it goes up a lot higher (over 100). But since it is a development LPAR, the response time is acceptable. The production WAS LPARs have less regions and more storage and demand paging is zero. We are also using 64-bit WAS and large pages (LFAREA) for production. Regards, Mark -- Mark Zelden - Zelden Consulting Services - z/OS, OS/390 and MVS mailto:m...@mzelden.com Mark's MVS Utilities: http://www.mzelden.com/mvsutil.html Systems Programming expert at http://expertanswercenter.techtarget.com/ -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Acceptable paging
Maybe you can explain what MQ and broker issues you have and why you think this is related to paging and/or the page configuration. I know the broker uses a lot of virtual storage, which indeed requires additional paging space, so the first check you can do is check how full your paging configuration is. Do the 'D ASM' command and check if each page dataset is well below 30% full. Kees. Lopez, Sharon sharon.lo...@nc.gov wrote in message news:239ab4569639f44abf2edc7f90f528b1f2b...@ncwwditmxmbx34.ad.ncmail.. . We have a z/10 and this test lpar had 11G of real memory. IBM is loaning us 16G of real memory so now we have 27G of real memory. We had 3 mod-9 page datasets and after message broker started we had to add 4 additional mod-9 page datasets. We do have IMS, CICS, and DB2 running on this system. Our tuning guy retired last year and none of us left understand how this is supposed to work. Thanks. -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of Lizette Koehler Sent: Wednesday, February 08, 2012 9:11 AM To: IBM-MAIN@bama.ua.edu Subject: Re: Acceptable paging Can someone please help me determine what the 'acceptable paging' should be on a system? We are having issues with MQ and broker. Thanks. Could you describe your environment a little? CPU(s) (z10, z9 , etc) Amount of dasd space allocated to Page Datasets? Number of Page Datasets Real Memory available Number of MQ and Brokers? Connection types (IMS, CICS, Distributed systems) Many elements can play into tuning a system. This will help in answering the question. Now, when you ask about acceptable paging in relationship to MQ, is anything else in play? CICS, IMS, DB2, heavy batch, etc. or is MQ the only thing running on that system? Or perhaps I misunderstood your acceptable paging in relationship to MQ. Thanks Lizette -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN E-mail correspondence to and from this address may be subject to the North Carolina Public Records Law and may be disclosed to third parties by an authorized state official. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.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...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Acceptable paging
Correction: *local*. Maybe you can explain what MQ and broker issues you have and why you think this is related to paging and/or the page configuration. I know the broker uses a lot of virtual storage, which indeed requires additional paging space, so the first check you can do is check how full your paging configuration is. Do the 'D ASM' command and check if each *local* page dataset is well below 30% full. Kees. Lopez, Sharon sharon.lo...@nc.gov wrote in message news:239ab4569639f44abf2edc7f90f528b1f2b...@ncwwditmxmbx34.ad.ncmail.. . We have a z/10 and this test lpar had 11G of real memory. IBM is loaning us 16G of real memory so now we have 27G of real memory. We had 3 mod-9 page datasets and after message broker started we had to add 4 additional mod-9 page datasets. We do have IMS, CICS, and DB2 running on this system. Our tuning guy retired last year and none of us left understand how this is supposed to work. Thanks. -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of Lizette Koehler Sent: Wednesday, February 08, 2012 9:11 AM To: IBM-MAIN@bama.ua.edu Subject: Re: Acceptable paging Can someone please help me determine what the 'acceptable paging' should be on a system? We are having issues with MQ and broker. Thanks. Could you describe your environment a little? CPU(s) (z10, z9 , etc) Amount of dasd space allocated to Page Datasets? Number of Page Datasets Real Memory available Number of MQ and Brokers? Connection types (IMS, CICS, Distributed systems) Many elements can play into tuning a system. This will help in answering the question. Now, when you ask about acceptable paging in relationship to MQ, is anything else in play? CICS, IMS, DB2, heavy batch, etc. or is MQ the only thing running on that system? Or perhaps I misunderstood your acceptable paging in relationship to MQ. Thanks Lizette -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN E-mail correspondence to and from this address may be subject to the North Carolina Public Records Law and may be disclosed to third parties by an authorized state official. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.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...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Acceptable paging
Demand paging is anathema to multi-user address spaces (e.g. CICS/IMS). It is somewhat less problematic for single user address spaces. (TSO, BATCH) and in-between for DB2. e.g. it may be perfectly acceptable to have a demand paging rate of 100 if CICS/IMS demand paging is zero. Side trip to the old days. I recall demand paging rates in the multi-hundreds along with swaps in the 80's. That being said, much research has already been done in this area. CMG (www.cmg.org) papers from the 80's and 90's will have a wealth of information about paging and its impacts on various workloads. SHARE (www.share.org) papers of the same vintage will also contain much useful information. Paging is not, by itself, a bad thing. When it becomes bad, as Mark said, is when the phone starts ringing or SLA's are not being met. The answer to your original question Can someone please help me determine what the 'acceptable paging should be on a system? is it depends. BTW, there are possibly other reasons for poor MQ and WAS performance unrelated to paging. HTH, -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Acceptable paging
In 239ab4569639f44abf2edc7f90f528b1f2b...@ncwwditmxmbx34.ad.ncmail, on 02/08/2012 at 02:01 PM, Lopez, Sharon sharon.lo...@nc.gov said: Can someone please help me determine what the 'acceptable paging' should be on a system? There is no one size fits all. We are having issues with MQ and broker. If the paging is causing unacceptable MQ performance then the paging rate is not acceptable *on your system*, even if a much higher paging rate is causing no problems on someone else's system. What are the data telling you? -- Shmuel (Seymour J.) Metz, SysProg and JOAT ISO position; see http://patriot.net/~shmuel/resume/brief.html 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...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Acceptable paging
The number and size of paging volumes set an upper limit to the amount of virtual storage that can be in use at any one time across all address spaces. This alone does not impact paging rates. Only the amount of real storage affects paging rates. (Note: this may not be 100% technically true, but it is close enough for this context.) A demand page in event can cause a full stop in the requesting address space as a main storage frame is made available (perhaps driving a page out to DASD operation) and the desired page is fetched from DASD. Even with very fast DASD with 100% cache hit rates, this is pretty slow. The amount of real memory determines how often a demand paging operation occurs. Ideally, that should average zero. As a practical matter, an overall system demand page rate of one or two pages per second can mean that you have a good balance between main storage and workload demand, but there isn't much room to grow. In my mind, 'acceptable paging' is when service levels are being met. Usually that means that there is little or no time spent waiting on demand paging operations. Unacceptable paging is when service levels are not being met AND the cause is diagnosed as waiting on page fault resolution. OBTW - I've found the following terms to be useful: A 'page' is 4096 bytes of virtual storage. A 'frame' is 4096 bytes of main storage. A 'slot' is 4096 bytes of DASD storage. A page always consumes a slot, but consumes a frame only when 'in'. HTH and good luck. -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of Lopez, Sharon Sent: Wednesday, February 08, 2012 8:26 AM To: IBM-MAIN@bama.ua.edu Subject: Re: Acceptable paging We have a z/10 and this test lpar had 11G of real memory. IBM is loaning us 16G of real memory so now we have 27G of real memory. We had 3 mod-9 page datasets and after message broker started we had to add 4 additional mod-9 page datasets. We do have IMS, CICS, and DB2 running on this system. Our tuning guy retired last year and none of us left understand how this is supposed to work. Thanks. -Original Message- From: IBM Mainframe Discussion List [mailto:IBM-MAIN@bama.ua.edu] On Behalf Of Lizette Koehler Sent: Wednesday, February 08, 2012 9:11 AM To: IBM-MAIN@bama.ua.edu Subject: Re: Acceptable paging Can someone please help me determine what the 'acceptable paging' should be on a system? We are having issues with MQ and broker. Thanks. Could you describe your environment a little? CPU(s) (z10, z9 , etc) Amount of dasd space allocated to Page Datasets? Number of Page Datasets Real Memory available Number of MQ and Brokers? Connection types (IMS, CICS, Distributed systems) Many elements can play into tuning a system. This will help in answering the question. Now, when you ask about acceptable paging in relationship to MQ, is anything else in play? CICS, IMS, DB2, heavy batch, etc. or is MQ the only thing running on that system? Or perhaps I misunderstood your acceptable paging in relationship to MQ. Thanks Lizette -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN E-mail correspondence to and from this address may be subject to the North Carolina Public Records Law and may be disclosed to third parties by an authorized state official. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN NOTICE: This electronic mail message and any files transmitted with it are intended exclusively for the individual or entity to which it is addressed. The message, together with any attachment, may contain confidential and/or privileged information. Any unauthorized review, use, printing, saving, copying, disclosure or distribution is strictly prohibited. If you have received this message in error, please immediately advise the sender by reply email and delete all copies. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Acceptable paging
In 1910aea19cd2554fb59403184ebe4381044b40a...@mmoexchmbs01.jhacorp.com, on 02/08/2012 at 04:49 PM, Hal Merritt hmerr...@jackhenry.com said: A demand page in event can cause a full stop in the requesting address space Address space or only task? A page always consumes a slot, No. An unreferenced page does not consume a slot. -- Shmuel (Seymour J.) Metz, SysProg and JOAT ISO position; see http://patriot.net/~shmuel/resume/brief.html 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...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Acceptable paging
I don't know why, but anyhow back in MVS internals. There's the PART, the SART and the Frame Allocation Resource Table. What no FART? Not PC! Sorry... In a message dated 2/8/2012 8:44:47 P.M. Central Standard Time, shmuel+ibm-m...@patriot.net writes: No. An unreferenced page does not consume a slot. -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
Re: Acceptable paging
Maybe you can explain what MQ and broker issues you have and why you think this is related to paging and/or the page configuration. I agree with Kees. What exactly is your 'issue' with 'the broker'? (Assuming that MQS is just the innocent bystander here.) I know the broker uses a lot of virtual storage, which indeed requires additional paging space, so the first check you can do is check how full your paging configuration is. 'The broker' uses a lot of shared storage (SHRLIBRGNSIZE), and it has a bad habit of polling for work (i.e., in all those numbered address spaces' timers pop at the same time), generating artificial 'loads' that WLM cannot keep up with. In our production environment 'the brokers' generally have a PI in the double digits (yes, double digit PI, sometimes up to triple digits - a bad clicker application NOT designed for resource sharing, i.e. z/OS). From what I understand, in the newest release they also 'require' a lot of 64bit storage, possibly due to some sort of websphere application server hidden in there (but I may be wrong on this). I also have the sneaking suspicion (which I am currently also investigating) that 'the broker' goes out and gets itself a lot of (virtual) storage, touches each and every of those pages and then never uses them. Which means they become slots out on AUX, or rather, they become dead weight. What's a few more Gig of storage to a clicker, anyway? If your issue is 'only' paging, then do a D ASM and add local page space until % usage is way below 30%. Make sure that you have your PAGTOTL in ieasys sufficiently high, as that can only be changed via IPL. Barbara -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN