Manuel Bouyer said:

>On Fri, Mar 01, 2002 at 02:02:21PM -0800, Andrew Gillham wrote:
>> On Fri, Mar 01, 2002 at 02:27:16PM -0700, Rick Kelly wrote:
>> > 
>> > And that is not really a high load by any stretch of the imagination on
>> > any system. 
>> 
>> As a measure of "utilization" it is high for an idle system.  Thinking about
>> "work load" it is not high at all.
>> 
>> An essentially idle system should not be sitting at 0.54 load.  That means
>> that half the time a process is running.  While that doesn't directly
>> translate to 50% cpu utilization, it seems high.
>
>Well, it depends on the period of time on which the measure has been done.
>Processes waiting on I/O are also accounted in the load average in one
>way or another I guess ...

We really don't know what was running on that machine.

I notice that the load averages from top will be slightly lower than "w" or
uptime at any given time.

-- 
Rick Kelly  rmk@rmkhome.com  www.rmkhome.com