Skip site navigation (1)Skip section navigation (2)
Date:        Sat, 4 Mar 2000 19:59:41 +0100
From:      Erik Trulsson <ertr1013@student.csd.uu.se>
To:        freebsd-questions@FreeBSD.ORG
Subject:   Re: Uptime/Load Averages
Message-ID:  <20000304195940.A337@student.csd.uu.se>
In-Reply-To: <4.1.20000304120821.0094f990@mail.udel.edu>; from papalia@udel.edu on Sat, Mar 04, 2000 at 12:10:39PM -0500
References:  <4.1.20000304120821.0094f990@mail.udel.edu>

next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, Mar 04, 2000 at 12:10:39PM -0500, John wrote:
> Hi all -
> 
> In an ever-present quest to learn and understand, I was trying to learn
> more about "load averages" as shown in uptime.  I read a quite extensive
> discussion in the archives over how the load average is *calculated*, but
> not exactly what it's saying.  I guess I'm wondering: is it an absolute
> scale? Is the min 0 and the max 100 or some other number?  Is a load avg of
> 0.1 good while a load avg of 1.0 is bad? Or is it not that cut and dry?
> What does knowing the load avg actually *tell* me.

The "load" is the number of processes that want to run at any given time.
Thus a load average of 0.0 means that no processes wanted to run while a
load average of 3.2 means that on average 3.2 processes wanted to run.
Thus the minimum is 0 and there is no max.
Anything below 1.0 is good. (Since that means that all processes got all the
CPU-time they wanted.)
A load above, say, 3.0 over a long time means that you probably need a
faster machine.
(If you have multiple CPUs you can multiply all the numbers above with the
number of CPUs.)





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




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000304195940.A337>