Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!ulowell!hen.ulowell.edu!arosen
From: arosen@hen.ulowell.edu (MFHorn)
Newsgroups: comp.sys.sequent
Subject: Dynix load average (was Re: passwd hashing)
Message-ID: <AROSEN.89Apr28004409@hen.ulowell.edu>
Date: 28 Apr 89 04:44:09 GMT
References: <2470@helios.ee.lbl.gov> <67727@pyramid.pyramid.com>
Sender: news@swan.ulowell.edu
Organization: University of Lowell, CS Dept
Lines: 24
In-reply-to: csg@pyramid.pyramid.com's message of 26 Apr 89 03:41:22 GMT

In article <67727@pyramid.pyramid.com> csg@pyramid.pyramid.com (Carl S. Gutekunst) writes:
   In article <2470@helios.ee.lbl.gov> Jef Poskanzer <jef@helios.ee.lbl.gov> writes:
   >(I haven't tried it on a heavily loaded system yet because I
   >haven't managed to get the load average above 1.6...)

   There's a reason for that. Dynix divides the load average by the number of
   CPUs you have. If uptime(1) displays 1.6, and you have four CPUs, then the
   load average is really 6.4.

If uptime(1) says "1.6" on a 4-CPU system, the load would really be 6.4
*on a single processor system*  If Dynix says the load is 1.6, it really
is 1.6 .  The load average is very relative to the machine you're working
on, though.

System               Memory  # users  'load average'  Delay in response time

20-CPU Balance 21K   22 Meg  50+        0.8           none
VAX 11/750            8 Meg  28        37.69          minutes per keystroke
MV/15000 (~12 Mips)  16 Meg   1        15+            none

Load averages can be very misleading, and should be looked at with respect
to the load average _on the same system_ under varying conditions.  It is
"the average number of jobs in the run queue over the last 1, 5 and 15
minutes", nothing more.