From: Brewer, Edward (BREWERE@OD.NIH.GOV)
Date: Tue Jan 28 2003 - 10:41:02 EST
Admins,
I ran across this statement in the online manual for 5.1a
link:
http://www.tru64unix.compaq.com/docs/base_doc/DOCUMENTATION/V51A_HTML/ARH9GB
TE/TITLE.HTM
The uptime command shows how long a system has been running and the load
average. The load average counts the jobs that are waiting for disk I/O, and
applications whose priorities have been changed with either the nice or the
renice command. The load average numbers give the average number of jobs in
the run queue for the last 5 seconds, the last 30 seconds, and the last 60
seconds.
collect states from the man pages:
RUNQ Number of processes in the run queue.
AVG5,30,60 Load average over the last 5, 30, and 60 seconds. (note
assume that this load average is the same as uptime)
If I examine the load averages and run que values versus the CPU time for a
whole day on one of my systems
User System Idle Wait
19.77067 7.421473 72.77658 0.027797
RunQue Load Avg 5 30 60
0.706741 2.535375 2.357224 2.331411
Notes: these values are averages from a single day collection. I haved
added all of the values and divided by the number of collections (30 second
interval)
The first line in the online manual claims that load average counts jobs
waiting on disk I/O and ones being niced. Then the second statement claims
that the values are the number of jobs in the run que for last 5, 30 and 60
seconds. If that is so, then why do the values differ in collect. My
average RunQue is .70, which I think means that for the day, I had an
average of close to one job running per second. My load average for 5/30/60
is close to 2.5 or 2.3. Does this mean that 1.5 jobs are waiting on
resources or being niced, or that I really had 2.5 jobs running per second.
Or is this a case of apples and oranges?
TIA,
Lee Brewer
This archive was generated by hypermail 2.1.7 : Sat Apr 12 2008 - 10:49:05 EDT