Date: Sun, 25 Mar 2001 13:23:18 -0600 (CST) From: David Scheidt <dscheidt@tumbolia.com> To: j mckitrick <jcm@FreeBSD-uk.eu.org> Cc: freebsd-questions@freebsd.org Subject: Re: interpreting 'load' statistics Message-ID: <Pine.BSF.4.21.0103251309270.43732-100000@shell-3.enteract.com> In-Reply-To: <20010325181304.A31661@dogma.freebsd-uk.eu.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, 25 Mar 2001, j mckitrick wrote: : :When you run 'w' or 'uptime', how can you interpret the load statistics? I :always thought under 1 was okay. I read that somewhere but now I'm not so :sure. : It depends on a bunch of things. The load averages are the average number of jobs in the run queue over the last 1 minute, 5 minutes, and 15 minutes. Depending on what the jobs are, there may not be a noticable slow down for other processes on the system. If you have a bunch of processes that are banging on the same disk, they're going to spend a fair amount of time blocked on disk I/O, so they'll show up as runnable, but other processes may not be effected by those. If you've got processes blocking because they're waiting for swap, everything is going to notice. In general a system with a load of less than (2 * number of CPUS) is probably doing alright, if the throughput of whatever it's doing is good enough. Above that, and I'd at least work out why. I might not do anything if the box were performing okay. I've run machines that have had loads above 10, all the time, which were doing their thing fast enough not to worry about. I've had others that fell apart and starting thrashing with a load of 2. David Scheidt -- dscheidt@tumbolia.com Bipedalism is only a fad. 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?Pine.BSF.4.21.0103251309270.43732-100000>