Date: Tue, 9 Dec 1997 11:50:21 -0800 (PST) From: Doug White <dwhite@gdi.uoregon.edu> To: Justin Ashworth <ashworth@cs.montana.edu> Cc: questions@FreeBSD.ORG Subject: Re: Uptime Message-ID: <Pine.BSF.3.96.971209114440.18119K-100000@gdi.uoregon.edu> In-Reply-To: <Pine.OSF.3.95.971209105715.7216A-100000@esus.cs.montana.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 9 Dec 1997, Justin Ashworth wrote: > What all goes into the load averages displayed by the uptime command? As > far as I can tell, the uptime has no theoretical upper limit, but there is > a certain max that it hits where simple processes would take 50+ times as > long to run. Does this value scale with the processor? In other words, > would the uptime on a 486-66 equal the uptime on a Pentium-120 running the > same processes? The CPU load average is a mess of different things, including number of processes waiting for disk I/O and other stuff. I've tried to find the code that computes this and haven't had any luck. I think it ended up on -hackers a few days ago, check the mail archives. The number is useful as a relative measure of activity on that particular machine. You can't compare it to other machines. For instance, just writing mail and running rc564 the load averages are 1.52, 1.45, 1.42 on my p133. I've never checked the load average after doing some heavy work on this box. Now, if you looked at the load average on the main CIS computer last week, it was running about 19 and it was starting to loose interactive response. Right now it's 2.04, 1.73, 1.53 with 29 users online and is pretty snappy. It's a 3 processor SparcCenter 1000. The load averages are completely relative. Doug White | University of Oregon Internet: dwhite@resnet.uoregon.edu | Residence Networking Assistant http://gladstone.uoregon.edu/~dwhite | Computer Science Major
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.96.971209114440.18119K-100000>