Date: Mon, 18 Oct 2010 16:31:10 -0400 From: Lowell Gilbert <freebsd-questions-local@be-well.ilk.org> To: Martin Turgeon <freebsd@optiksecurite.com> Cc: freebsd-questions@freebsd.org Subject: Re: How is CPU usage calculated? Message-ID: <44pqv7tf8x.fsf@be-well.ilk.org> In-Reply-To: <4CBCA8D1.5040005@optiksecurite.com> (Martin Turgeon's message of "Mon, 18 Oct 2010 16:06:41 -0400") References: <20101015071719.1536f19e@mail.ose.nl> <4CBC97CB.80604@optiksecurite.com> <AANLkTimT_4D4THvszdDS=U0UKtScGgKZEN4A9e8BB2%2Bf@mail.gmail.com> <4CBCA8D1.5040005@optiksecurite.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Martin Turgeon <freebsd@optiksecurite.com> writes: > Le 2010-10-18 15:12, Adam Vande More a =E9crit : >> On Mon, Oct 18, 2010 at 1:54 PM, Martin Turgeon >> <freebsd@optiksecurite.com>wrote: >> >>> I didn't knew about -H to show individual threads, but my problem isn't >>> there. My problem is that the summary printed in the first lines doesn't >>> match the total of the process detailled under the summary. For example= , I >>> have a user CPU usage est 83.1% in the summary, but the only process wo= rth >>> mentionning in the list is mysqld that is taking 255.52% WCPU. That's f= ar >>> less than half the "CPU power" but 83% is far more. It's that difference >>> that I don't understand. >>> >>> How is this possible? >>> >> man 1 top >> >> 83.1% doesn't mean what you think it does. >> > I just reread it and it isn't clearer what is using the CPU so > much. Can you please give me a little more explanation? They are calculated over different time periods, averaged in different ways (that's they "weighted" part), and processes only show up until they exit.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?44pqv7tf8x.fsf>