Possible Duplicate:
How to understand the memory usage and load average in linux server
My hosting company provides "Load Average" stats when I login. There is no link to more information about them, and no explanation as to what they mean. I'm guessing they are a percentage, but I have no clue why there are three numbers, or what the percentage is actually out of (processor, memory, bandwidth, combination?), and is that this instant or over time? Is this a standard statistic that I should know about?
What are high / low / normal numbers?
Perhaps someone can shed some light on this for me.
Examples:
Load Averages: 0.60 0.60 0.55
The three numbers represent the number of processes, on average, that are ready to run but are blocked, over the last one, five, and fifteen minutes, respectively.
https://secure.wikimedia.org/wikipedia/en/wiki/Load_average
It varies very slightly from one unix-like OS to the next (e.g., different criteria for "waiting"), but they all work basically the same way.
Load averages measure the trend in CPU utilization by calculating all demand for the CPU over a period of time, which can be thought of as a moving average. The important point to notice is that the value decreases with age. The figures (which are read left to right) do not include any processes or threads waiting on I/O, networking, databases or anything else not demanding the CPU. It only focuses on what actively demands CPU time, which is entirely different from CPU percentage.
By using the function y == (2^11/2^((5 Log2[E])/(60 x))) with the parameters of the load-average bucket, a constant exponent, and the number of running/uninterruptible processes, an exponential decay function is produced. (EXP_1 / 1 min. average, EXP_5 / 5 min. average and EXP_15 / 15 min. average).
For example if there are four CPUs on a server and it reported a one-minute load average of 4.00, then this indicates the machine has been utilizing its processors perfectly during the past 60 seconds.