1

when I issue the vmstat command without any parameter I get this result:

# vmstat
procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu------
r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa st
7 117      0 719328 1251624 258624672    0    0   346    64    0    0  4  0 92  3  0

As you can see cpu idle is 92 percent. No matter how many time I issue the command, I get similar results.

But when I issue vmstat with a time interval parameter, cpu idle value falls down immediately:

# vmstat 5        
    procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu------
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa st
385  0      0 145380064 1457188 115708528    0    0   346    64    0    0  4  0 92  3  0
417  0      0 144694256 1457204 116388928    0    0     0    49 1122 528500 84  5 12  0  0
413  0      0 143960640 1457220 117117040    0    0     0    31 1127 529619 84  5 11  0  0
418  0      0 143224224 1457228 117850080    0    0     0    26 1114 530135 84  5 11  0  0
444  0      0 142483520 1457240 118587048    0    0     0    33 1112 530314 84  5 11  0  0
397  0      0 141820240 1457256 119245040    0    0     0    37 1114 531348 85  5 10  0  0

Here it is! Even if I get results in 5 secs interval, cpu idle value falls about 80 percent!

I can't imagine if vmstat itself consume this much CPU power ( server has 32x Inter Xeon X7550 CPUs each has two 2 cores and 2 threads).

Can someone give me a clue about that %80 Cpu power I loose?

ramazan polat
  • 148
  • 2
  • 10

1 Answers1

3

From the man page:

If no delay is specified, only one report is printed with the average values since boot.

You are comparing the average idle time since the machine is up (92%) with five second intervals (10-12%), which doesn't make sense.

Oliver
  • 5,973
  • 24
  • 33