The CPU idle is set to 99 while using vmstat, what does this mean

performancevmstat

I am confused with the CPU idle time is set to 99. Does this mean the CPU is idle not doing anything for 99% of the time?

#vmstate
procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu------
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa st
 0  0   2336  18512 263972 1048992   0    0    1     47   0    0  0  0  99  0  0

Best Answer

Please note, that on most unix systems first line of vmstat output is average since last reboot. Next lines are current statistics.

99% CPU Idle means that your server is not really heavy loaded. Which is good. Nothing to worry about :)

Related Solutions

How to Get Memory Usage Percentage with vmstat

What you're looking for is the output from "free":

$ free
             total       used       free     shared    buffers     cached
Mem:        775556     759456      16100          0      22132     592484
-/+ buffers/cache:     144840     630716
Swap:       500344      21972     478372

Here's a tour:

This is a box w/ 768MB of physical RAM and a 500344KB swap partition.

759456KB is "used" (second column, top row). Of this "used" memory, 23132KB is buffers (5th column, top row) and 592484KB (sixth column, top row) is cache, leaving 144840KB (2nd column, 2nd row) of physical memory that's being used by active processes.

When you consider that the memory used by buffers and cache could be used for processes, that leaves 630716KB (3rd column, 2nd row) of physical memory free.

The swap partition, as I said, is 500344KB (1st column, bottom row). 21972KB (2nd column, bottom row) of the swap file is in use, leaving 478372KB (3rd column, bottom row) free.

So, your definition of % free memory depends on whether you're counting buffers and cache or not, and whether you're counting swap or not.

That ought to give you enough to go on to calculate a percentage as you see fit.

Here's the output of top and free:

$ top -n1 | grep "used," ; free
Mem:    775556k total,   751472k used,    24084k free,    20776k buffers
Swap:   500344k total,    21972k used,   478372k free,   586648k cached

             total       used       free     shared    buffers     cached
Mem:        775556     751344      24212          0      20776     586648
-/+ buffers/cache:     143920     631636
Swap:       500344      21972     478372

You can see how top and free agree (albeit with a little difference-- this is a multi-user machine and the memory usage of the top and free programs are reflected in there).

Add a "-s" when you call vmstate, and you will see much the same output, just in a different form:

$ vmstat -s ; free
       775556  total memory
       759920  used memory
       674680  active memory
        18440  inactive memory
        15636  free memory
        21892  buffer memory
       594372  swap cache
       500344  total swap
        21972  used swap
       478372  free swap
       ... output truncated ...

             total       used       free     shared    buffers     cached
Mem:        775556     759920      15636          0      21892     594372
-/+ buffers/cache:     143656     631900
Swap:       500344      21972     478372

You can see that vmstat is showing all the same numbers as free.

Performance IO – Are Networks Now Faster Than Disks?

Here are some numbers that you are probably looking for, as quoted by Jeff Dean, a Google Fellow:

Numbers Everyone Should Know

L1 cache reference                             0.5 ns
Branch mispredict                              5 ns
L2 cache reference                             7 ns
Mutex lock/unlock                            100 ns (25)
Main memory reference                        100 ns
Compress 1K bytes with Zippy              10,000 ns (3,000)
Send 2K bytes over 1 Gbps network         20,000 ns
Read 1 MB sequentially from memory       250,000 ns
Round trip within same datacenter        500,000 ns
Disk seek                             10,000,000 ns
Read 1 MB sequentially from network   10,000,000 ns
Read 1 MB sequentially from disk      30,000,000 ns (20,000,000)
Send packet CA->Netherlands->CA      150,000,000 ns

It's from his presentation titled Designs, Lessons and Advice from Building Large Distributed Systems and you can get it here:

Dr Jeff Dean Keynote PDF or on slideshare.net

The talk was given at Large-Scale Distributed Systems and Middleware (LADIS) 2009.

Other Info

It's said that gcc -O4 emails your code to Jeff Dean for a rewrite.