From owner-freebsd-hackers  Fri May  3 10:36:10 2002
Delivered-To: freebsd-hackers@freebsd.org
Received: from tcpns.com (dsl-64-192-239-221.telocity.com [64.192.239.221])
	by hub.freebsd.org (Postfix) with ESMTP
	id 03A3D37B417; Fri,  3 May 2002 10:36:05 -0700 (PDT)
Received: from localhost (localhost [127.0.0.1])
	by tcpns.com (8.12.3/8.12.2) with ESMTP id g43Ha4eP076250;
	Fri, 3 May 2002 13:36:04 -0400 (EDT)
Date: Fri, 3 May 2002 13:36:04 -0400 (EDT)
From: Jason Borkowsky <jcborkow@tcpns.com>
To: John Baldwin <jhb@FreeBSD.org>
Cc: hackers@FreeBSD.org
Subject: RE: CPU context switching/load numbers
In-Reply-To: <XFMail.20020503103549.jhb@FreeBSD.org>
Message-ID: <Pine.BSF.4.44.0205031334390.75809-100000@bemused.tcpns.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-freebsd-hackers@FreeBSD.ORG
Precedence: bulk
List-ID: <freebsd-hackers.FreeBSD.ORG>
List-Archive: <http://docs.freebsd.org/mail/> (Web Archive)
List-Help: <mailto:majordomo@FreeBSD.ORG?subject=help> (List Instructions)
List-Subscribe: <mailto:majordomo@FreeBSD.ORG?subject=subscribe%20freebsd-hackers>
List-Unsubscribe: <mailto:majordomo@FreeBSD.ORG?subject=unsubscribe%20freebsd-hackers>
X-Loop: FreeBSD.ORG


> >> > Greetings! I have a FreeBSD-4.5 box that is a specialized server box. It
> >> > doesn't run any user processes and only runs a bunch of small, server
> >> > efficient processes.
> >> >
> >> > I have an inconsistency that I am trying to explain. When I do a "w"
> >> > command
> >> > on the box, I see this:
> >> >
> >> >  7:31PM  up 74 days, 39 mins, 1 user, load averages: 1.12, 0.94, 0.93
> >> >
> >> > This says I have a load of 1.12 over the past minute, or, for every
> >> > available CPU interval, I have 1.12 processes requesting the CPU.
> >>
> >> This last bit is where you go wrong.  The 1.12 is just for the minute prior
> >> to
> >> when you ran the command, it has no relation to any previous minutes.  Just
> >> cause it is 1.12 right now doesn't mean the average load for every minute is
> >> 1.12.
> >
> > But these numbers are over months...I have used an expect script to
> > periodically poll the load and vmstat, and save them off to a file. My
> > average load over a three month period is about 0.98, but the average CPU
> > idle time over the same 3 month period is about 85% idle.
>
> Yes, and when your script was running, it was runnable, right?  So it
> artifically inflated the last minute's load.

But the script polled only once every 5 minutes, just running an "uptime"
and cat'ing it to a file. There is no way this would have instantaneously
affected the 1 minute load average, no less the 5 and 15 minute load
averages.


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message