On Monday 28 March 2005 11:03, Gabor Szabo wrote: > One of the things which is not clear to me is that it seems there is a > total lack of connection between the average load and the CPU states. > Could someone explain why that would be ?
The load average is not in percentage. The load average numbers are the average number of processes waiting/using for CPU in the last 1, 5 and 15 minutes [remark: on Linux processes in the 'D' (uninterruptible sleep) are weirdly in this count also]. How to interpret this? Is a 1.25 a high or low number? Well, on a single CPU system this means it runs over its capacity because more than a single process could have used the CPU at that time frame but there was only one CPU was available. However, on a 4 CPU system this means they are barely utilized (and there may be other bottlenecks like memory, I/O etc.). > 10:50am up 59 days, 18:47, 3 users, load average: 1.25, 1.17, 1.27 > 125 processes: 124 sleeping, 1 running, 0 zombie, 0 stopped > CPU0 states: 17.0% user, 21.0% system, 0.0% nice, 60.0% idle > CPU1 states: 27.0% user, 11.0% system, 0.0% nice, 60.0% idle Your system shows 2 CPU's but I suspect you really have 1 Hyperthreaded CPU (do you?). In that case it looks like you are using roughly all your CPU capacity. > Mem: 2579260K av, 2518756K used, 60504K free, 0K shrd, 76608K buff > Swap: 522072K av, 0K used, 522072K free 2109908K cached Also memory seems to be good (no swaping at all yet). Looks to me like: "I want more CPU power" Hope it helps, -- Oron Peled Voice/Fax: +972-4-8228492 [EMAIL PROTECTED] http://www.actcom.co.il/~oron ICQ UIN: 16527398 "I love deadlines, especially the whooshing sound they make as they go by." -- Douglas Adams ================================================================= To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
