Re: CPU VP Issue
Posted in 2008
Topics: Versions, Editions & End-of-Life
Not a hundred percent sure, but theres a processor affinity setting that may be forcing only one cpu. Also, what threads are sitting ? Maybe the app is single threaded ? ----- Original Message ----- Subject: CPU VP Issue Date: Thu, May 15, 2008 20:28 From: "tgirsch" <tgirsch@NOSPAM.gmail.com> I've got a puzzling CPU VP issue that I'm hoping someone else here has > dealt with. I've got a dedicated database server running IDS 10.00.FC8 > on a four-core RHEL Opteron server. We've configured four CPU VPs, one > per core. There are almost always five to six threads sitting on the > ready queue, despite the fact that the system is generally > 60% idle time. > > This sparks the age-old debate: add more CPU VPs than physical processor > cores? Or is there something incorrectly configured about the _current_ > setup that's causing my CPU VPs to do less than they should / yield too > easily, etc.? > _______________________________________________ > Informix-list mailing list > Informix-list@iiug.org > http://www.iiug.org/mailman/listinfo/informix-list > >
Floyd Wellershaus wrote: > Not a hundred percent sure, but theres a processor affinity setting that > may be forcing only one cpu. > Also, what threads are sitting ? Maybe the app is single threaded ? > We haven't set processor affinity, although we've toyed with that idea. Not sure if it's even supported on RHEL 4. The app is decidedly NOT single-threaded. Usually, on the active queue, there's an even mix of sqlexec and kaio threads, although we do frequently see more kaio than sqlexec threads. The ready queue is mostly sqlexecs. It should be noted that this server is the "target" in a primary-target ER relationship, and is read-only from a client perspective. So we have lots of clients connecting to do read-only operations, and the only writes that go on are updates as a result of ER.