Xref: utzoo unix-pc.sources:401 comp.sys.att:7800
Path: utzoo!utgpu!watmath!uunet!cs.utexas.edu!rutgers!bpa!manta!brant
From: brant@manta.pha.pa.us (Brant Cheikes)
Newsgroups: unix-pc.sources,comp.sys.att
Subject: Re: SYSINFO utility for information on UNIX pc SLK lines (repost)
Summary: explanation for load avg figures
Keywords: sysinfo, slk, labels, information, repost, version2.1
Message-ID: <496@manta.pha.pa.us>
Date: 17 Oct 89 16:19:51 GMT
References: <986@icus.islp.ny.us>
Reply-To: brant@manta.pha.pa.us (Brant Cheikes)
Organization: Soul of the Gnu Machine, Philadelphia
Lines: 21

Lenny Tropiano writes:
>This version has the loadavgd (daemon) to report load avg results.  Now
>if someone could EXPLAIN to me the significance of the three numbers,
>other than the larger the number the more loaded the machine.

First, thanks to Lenny & the other folks who contributed to sysinfo.
It's a very handy little utility (BTW, I've replaced all previous
versions of sysinfo on osu-cis with the new version, and included
binaries to boot).

As far as the explanation goes, if the numbers have the same meaning
as in BSD, then they reflect the average over the last minute, 5
minutes, and 15 minutes, of processes eligible to run (in queue).
The one-minute figure gives you the strongest sense of dynamic system
load.  When the number climbs, more processes are contending for the
CPU, and thus turnaround time increases (so the system seems to get
slower).
-- 
Brant Cheikes
University of Pennsylvania, Department of Computer and Information Science
brant@manta.pha.pa.us, brant@linc.cis.upenn.edu, bpa!manta!brant