What is the “load average” on Linux server.

[bullz]

I have read couple of article on internet, but still I am confused.
Could someone please give me little explanation on load average and find the list of process which are consuming more on load average.

load average: 4.23, 4.84, 4.89

Points will be shared without fail :)

[Matti_Kurkela]

I found this very clear explanation:

http://blog.scoutapp.com/articles/2009/07/31/understanding-load-averages

If you're still confused after reading this, could you please explain in more detail what's confusing you?

> load average: 4.23, 4.84, 4.89

Your short-term load values are smaller than longer-term load values => the load seems to be decreasing at the moment, although this might not be reliable if your workload contains spikes of heavy load and periods of relative idleness.

The "top" command can be used to quickly find the biggest CPU users.

After you master "top", you might try "atop" too. It provides more information than regular "top": in addition to regular CPU load, it can also detect I/O workload. The latter is very useful in detecting I/O-related performance problems.

http://www.atoptool.nl/

MK

[Alzhy]

Bullz..

Load on LINUX is vastly different than "load" on HP-UX or other UNIX dialects.

Most Monitoring Tools (Commercial) still follow the same train of thought like what is mentioned in the URL suggested but it is different on Linux actually these days.

Your "LOAD" can go astronomically HIGH with your CPU virtually untaxed - as you will hopefully find out. Like, we had a server with an unused LVM VG whose PV "disappeared" and there were several running LVM Commands in the background (Unkillable - but nary a CPU use) -- and this WAS causing a server to register a load of over 20! (system is a 24 core server)... I traced this fact to an excellently written URL explaining "Load" on today's Linux -- I will post that URL once I locate it.

But it seems LOAD in Linux is no longer just that traditional guage of a CPU being loaded and processes getting backed up...


Cheers!

[dirk dierickx]

afaik, there is no difference in the meaning of these numbers across unix's, simply put it is a representation of processes in the queue.

in the man page on linux you'll find;

System load averages is the average number of processes that are either
in a runnable or uninterruptable state. A process in a runnable state is
either using the CPU or waiting to use the CPU. A process in
uninterruptable state is waiting for some I/O access, eg waiting for
disk. The averages are taken over the three time intervals.
Load averages are not normalized for the number of CPUs in a system, so
a load average of 1 means a single CPU system is loaded all the time
while on a 4 CPU system it means it was idle 75% of the time.


where linux differs is the way it calculates these values, where other unix's use clock-ticks, linux uses the HZ frequency (from wikipedia);

On Linux systems, the load-average is not calculated on each clock tick, but driven by a variable value that is based on the HZ frequency setting and tested on each clock tick.

[Alzhy]

Dirky -- it is different...

CPU interrupts are not factored in LOAD on most UNICES.... Not on REDHAT though...