There is a much lower limit to worry about than the one that prevents new works from starting. At around 50%, SVC dump will fail with 'ASM shortage'. This barrier has been discussed recently at SHARE. IBM agrees that SVCDUMP's ASM calculation as implemented is too strict, but it still carries the day. With no SVC dumps possible, many would consider a system hobbled.
. . JO.Skip Robinson SCE Infrastructure Technology Services Electric Dragon Team Paddler SHARE MVS Program Co-Manager 626-302-7535 Office 323-715-0595 Mobile jo.skip.robin...@sce.com From: Mark Zelden <m...@mzelden.com> To: IBM-MAIN@bama.ua.edu Date: 02/02/2012 01:02 PM Subject: Re: Very Lage Page Datasets (was ASM and HiperPAV) Sent by: IBM Mainframe Discussion List <IBM-MAIN@bama.ua.edu> On Thu, 2 Feb 2012 13:42:45 -0500, Jim Mulder <d10j...@us.ibm.com> wrote: >> >I wonder what if any studies on this have been done in the lab. >> >It would be nice if an IBM performance expert like Kathy Walsh >> >could weigh in. Thanks for jumping in Jim! I hoped you or Peter would eventually to clear up some of this FUD. > > The last performance studies I remember for paging were >around the time of MVS XA/SP2.2.0. Very little has been done >in the area of paging performance since then except for the >PAV stuff to allow two concurrent operations to a page data set. > Even though "zero" demand paging is best and what many shops strive / configure for these days, I think it's time for a new study based on modern architecture and what the OS now does. > > The page data set selection algorithm considers service time for >the devices, but not the full percentage. One could argue that >the full percentage should be considered, since it affects the >likihood of finding contiguous slots, and the CPU time to find >available slots, but that is not how it currently works. I'm going to take a giant leap of faith here and say there is zero to negligible performance impact in having a farm of 3390-27 local page datasets running at 50% full compared to having them at 30% (or less). Maybe 69% is fine too. No higher - we certainly don't want to hit that 70% MCCASMT1 threshold! :-) Regards, Mark ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
