Date: Sun, 18 Jan 2004 00:47:49 +0000 From: Colin Percival <colin.percival@wadham.ox.ac.uk> To: Jeremy Faulkner <gldisater@gldis.ca> Cc: freebsd-chat@freebsd.org Subject: Re: Good BSD/Linux Article (somewhat off-topic) Message-ID: <6.0.1.1.1.20040118004459.03fe6c10@imap.sfu.ca> In-Reply-To: <4009D71E.3020209@gldis.ca> References: <6.0.1.1.1.20040116175159.03f4dd48@imap.sfu.ca> <Pine.NEB.3.96L.1040117185613.22159B-100000@fledge.watson.org> <6.0.1.1.1.20040118000417.02bbee70@imap.sfu.ca> <4009D71E.3020209@gldis.ca>
next in thread | previous in thread | raw e-mail | index | archive | help
At 00:45 18/01/2004, Jeremy Faulkner wrote: >Colin Percival wrote: > > Actually, this raises an interesting point -- if >>1. There is a significant amount of network traffic, >>2. There is memory pressure, and >>3. There are several runnable processes, >>it might be a good idea to give scheduling priority to the oldest >>process, in the hope that it will complete and free its memory. >>Colin Percival > >dnetc and seti would be the oldest process on some machines. So making >this a mandatory setting would be counter productive. You're absolutely right -- s/oldest process/oldest process which started within the past 5 seconds/ would probably be more appropriate. (And, for bonus points, we could make this behaviour depend upon a global variable which would be set by network drivers if they detected the string "slashdot" within any incoming packets...) Colin Percival
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?6.0.1.1.1.20040118004459.03fe6c10>