Hi Marcy, Forgive my ignorance but what is the "vm.swappiness setting"?
Thank You, Terry Martin Lockheed Martin - Information Technology z/OS & z/VM Systems - Performance and Tuning Cell - 443 632-4191 Work - 410 786-0386 [email protected] -----Original Message----- From: The IBM z/VM Operating System [mailto:[email protected]] On Behalf Of Marcy Cortes Sent: Wednesday, January 28, 2009 2:34 PM To: [email protected] Subject: Re: Linux Guest 'swapping' That's been my experience as well. You may want to look at the vm.swappiness setting. We changed it from 60 to 20 because some of the applications swap space seemed to just grow and grow and grow and CPU would start to creep up too. Dropping to 20 removed all the issues. I'm not sure why and we never did get a good answer from anyone... The application in question here might have a memory leak too. That seems to be pretty common in vendor apps as well as your own java stuff. Marcy "This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose, or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation." -----Original Message----- From: The IBM z/VM Operating System [mailto:[email protected]] On Behalf Of Mark Post Sent: Wednesday, January 28, 2009 9:26 AM To: [email protected] Subject: Re: [IBMVM] Linux Guest 'swapping' >>> On 1/28/2009 at 9:57 AM, Robert J Brenneman <[email protected]> wrote: > Just a guess till the experts chime in: > > Linux disk I/O activity requires more CPU time than traditional Z > Operating systems - so when one guest starts driving 5000 I/O ops per > second to the swap device ( FBA mode vdisk in my case ) that in itself > consumes a big chunk of CPU. Then there's the additional time spent in > the linux kernel itself deciding what needs to go out to swap and what > needs to come back in. My experience isn't that it's the CPU cost of the paging I/O. It's more that Linux, like any other virtual storage operating system, can get into a thrashing mode, and all the kernel is doing is lots and lots of memory management. The system runs at 100% busy, and nothing gets done. Mark Post
