I don't know that much about other OSes, but at least for recent Linux (>=2.6 & RHEL3's hybrid 2.4/2.6 kernel), wio cpu is a very important metric and the kernel can often spend significant amounts of time there. It is one of the metrics that we pay the most attention to on the webfrontend as it usually indicates either a kernel problem (RHEL3) or at the very least an inefficient use of resources since the CPU is wasting time waiting for I/O and not doing real work.
Maybe ganglia can use a more flexible way to create the canned graphs and then have OS-specific profiles that it can display them automatically, via the conf.php or UI? ~Jason On Fri, 2005-03-25 at 17:52 -0800, Brooks Davis wrote: > CPU counters are another issue. Most systems support reporting user, > nice, system, interrupt, and idle usage, but ganglia also reports wio and > sintr. These are interesting values, but they aren't always available. > Presumably, other systems support other counts that are meaningful > there. It seems like what we need here is the ability to say, "this set > of named values adds to 100% of this resource" with corresponding > support in both gmetad and the web frontend. There might also need to > be a way to sort those values into groups to allow somewhat useful > summary graphs for mixed clusters. Off the top of my head, grouping CPU > into user, system, and idle possibly plus interrupt would give use > pretty graphs for just about anything. > > Any thoughts on these issues? > > -- Brooks > -- /------------------------------------------------------------------\ | Jason A. Smith Email: [EMAIL PROTECTED] | | Atlas Computing Facility, Bldg. 510M Phone: (631)344-4226 | | Brookhaven National Lab, P.O. Box 5000 Fax: (631)344-7616 | | Upton, NY 11973-5000 | \------------------------------------------------------------------/
