0

i have a problem with the load average on server, I'm using on server When I check the processor - with "top" command I see that :

top - 04:27:37 up 39 min,  1 user,  load average: 0.78, 0.77, 0.77
Tasks: 190 total,   1 running, 188 sleeping,   0 stopped,   1 zombie
Cpu(s):  0.3%us,  0.1%sy,  0.0%ni, 99.6%id,  0.0%wa,  0.0%hi,  0.0%si,  0.0%st
Mem:   7136868k total,  1504092k used,  5632776k free,    52552k buffers
Swap:        0k total,        0k used,        0k free,   602524k cached

it's the same all the time.. cpu not working but load average is high! why? Where load average was taken?

thanks

1
  • 1
    0.77 isn't high, even for only one core
    – theist
    Jan 9, 2012 at 19:08

2 Answers 2

1

Your load average is 0.78, which isn't really high. How many processors do you have in there? If your load average is less than the number of cores in there, don't worry about it.

3
  • i have 7 processors (Intel Xeon E5410 2.33GHz) but i don't understand, if all CPU(s) row equal to zero why load average equal to 0.78 ? The average load should be much smaller. also After 30 minutes the load average stay on 0.78-0.77 What could be not ok ?
    – user106427
    Jan 9, 2012 at 15:39
  • easy way to think is.. a load of 1 for 1 cpu system is 100% if your OS sees 8 cpus.. a load of .78 is nothing.. a load of 8 is something to check into
    – Mike
    Jan 9, 2012 at 16:16
  • The load average is the number of processes waiting to run. So a .78 load average means that one process was waiting for about 3/4 of the sample time. Since there are 8 processors to wait on... not a big deal.
    – Bill Weiss
    Jan 9, 2012 at 22:37
0

I wouldn't say that 0.78 is particularly high for a multi CPU setup. But it depends. Are you suffering for some reason due to this load? Is the system running bad? Sometimes you can have very little CPU "workload" while you still see a higher load average, could be caused by a lot of things, but in personal experience it has often been related to I/O. Processes reading or writing a lot from disks and haven't always been able to keep up, thus have to wait a little bit and then affecting the load average.

A good and simple explanation of Linux load average can be read at:
http://blog.scoutapp.com/articles/2009/07/31/understanding-load-averages

You must log in to answer this question.