I would check with Elpida to confirm that. We don't know if the said customer is at the right release of products, operating system, or has all fixes implemented. Seems like an awfully big hole in what all collectively think we seem to know.
-----Original Message----- From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of Pinnacle Sent: Sunday, August 15, 2010 Sunday 11:56 AM To: [email protected] Subject: Re: Anyone Using z10 Large Pages in anger? ----- Original Message ----- From: "Mark Zelden" <[email protected]> Newsgroups: bit.listserv.ibm-main Sent: Saturday, August 14, 2010 10:27 AM Subject: Re: Anyone Using z10 Large Pages in anger? > On Sat, 14 Aug 2010 10:24:02 +0100, Graham Harris <[email protected]> > wrote: > >>Is anyone actively using z10 (or z196!!) Large Pages in any significant >>way >>yet? >> > > Yes. On my clients WebSphere LPARs. First we used a small LFAREA and > migrated some WAS regions to 64-bit. All was fine. Then on a weekend we > allocated about 40G more to LFAREA and IPLed with the plan of converting > more regions during the week. Much to our chagrin, Monday morning > was a nightmare because paging went through the roof and thousands > of enclaves were backed up in the system and couldn't be processed. > > It turns out the support didn't work as documented and 1M pages weren't > really used for 4K when there was a shortage! After we opened a Sev 1 > PMR with IBM during the problem IBM said something like "well, this isn't > really supported". > > Huh?!!! This was z/OS 1.10 and I thought the support was z/OS 1.9 > with PTFs. :-) The SHARE presentations I was at and documentation I > had > read sure made it sound like it was supported! > > During a tech call with IBM they admitted that the support wasn't working > as intended (?) and the LFAREA was really just "stuck" there all for 1M > pages and couldn't be used to back 4K page requests. So in essence we > lost 40G of real memory when we IPLed. > > We converted a some additional regions to 64-bit over a few hours around > lunch > time to limp through the day without an outage, then backed out the > changes > that evening with an IPL along with any WAS regions that were converted > to 64-bit. > > We are still using large pages but have been doing it in steps over the > last > 6 months (IPL with larger LFAREA, convert more WAS regions to 64-bit). > > We are z/OS 1.11 now and I honestly can't say if anything was changed in > the code or if it works yet. The person closest to this works on the > WebSphere team and I haven't asked. Maybe Elpida is lurking and > can say. > > This weekend my client is doing a memory upgrade and also increasing > LFAREA > in one of the LPARs from around 64G to 80G (out of about 170G total). > Wow. Hard to believe Elpida would have shipped that crap. Thanks for the heads up on Large Page non-support. Regards, Tom Conley ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
