Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

Possible Duplicate:
How to understand the memory usage and load average in linux server

I have ready many articles that try to explain load average. To date none have ever explained it well enough that my thick brain could understand it.

Please could someone explain what load average means, in the simplest terms possible, so that anyone can understand it and make use of it.

share|improve this question
    
have a look at the answers to this question –  Iain Mar 25 '11 at 19:26
    
    
The way the question is asked here makes the entire subject much easier to understand than on the other threads IMHO –  Justin Vincent Mar 27 '11 at 4:10
add comment

marked as duplicate by Scott Pack, Jim B, sysadmin1138 Mar 25 '11 at 19:36

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

2 Answers

There's a lot of variation in how load averages are derived from system to system. That said, and without going into nitty-gritty detail, a decent analogy or high-level view of load average can be envisioned like this:

A 1.0 load average is about equivalent to one core of a processor working at 100%.

If you have 4 cores, and are running a 4.00 load average, your processing power is running full-steam. So, if your load average exceeds the number of cores you have, CPU requests are being delayed because the load is too high.

The key difference between load average and CPU usage is the total demand placed on the processor. Your CPU can be running at 100% and still have requests waiting to be fulfilled, for example.

share|improve this answer
    
Great answer thanks. The way you explain makes it very understandable (unlike other threads) –  Justin Vincent Mar 27 '11 at 4:11
    
@Justin I've been told that I should also note that I/O wait is added to that same queue, so be aware that it's possible for an overloaded storage device to also bump up your load average –  Hyppy Mar 27 '11 at 4:37
add comment

I'll take a stab at it:

Think of the load average as your "to-do" list and imagine you can only do one task at a time. So, if you have between 0-1 things on your to-do list, you're okay and maybe even have a little free time every now and then. If your to-do list stays right at or under 1, you're busy, but not overwhelmed. If things start piling up and the length of your to-do list grows past 1, you're overwhelmed, have no free time, and unable to get things done in a timely fashion. In short, you have too much to do.

These numbers illustrate a single-core system.

Link to a much better article on this:

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

share|improve this answer
add comment

Not the answer you're looking for? Browse other questions tagged or ask your own question.