From owner-freebsd-arch@FreeBSD.ORG Fri Aug 1 07:41:18 2003 Return-Path: Delivered-To: freebsd-arch@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 4EFA337B401 for ; Fri, 1 Aug 2003 07:41:18 -0700 (PDT) Received: from dan.emsphone.com (dan.emsphone.com [199.67.51.101]) by mx1.FreeBSD.org (Postfix) with ESMTP id 6975343F85 for ; Fri, 1 Aug 2003 07:41:17 -0700 (PDT) (envelope-from dan@dan.emsphone.com) Received: (from dan@localhost) by dan.emsphone.com (8.12.9/8.12.9) id h71Ef9DI049878; Fri, 1 Aug 2003 09:41:09 -0500 (CDT) (envelope-from dan) Date: Fri, 1 Aug 2003 09:41:09 -0500 From: Dan Nelson To: Poul-Henning Kamp Message-ID: <20030801144109.GG13080@dan.emsphone.com> References: <7379.1059732120@critter.freebsd.dk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <7379.1059732120@critter.freebsd.dk> X-OS: FreeBSD 5.1-CURRENT X-message-flag: Outlook Error User-Agent: Mutt/1.5.4i cc: arch@freebsd.org cc: Dag-Erling Smorgrav cc: Peter Jeremy Subject: Re: headsup: swap_pager.c X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 01 Aug 2003 14:41:18 -0000 In the last episode (Aug 01), Poul-Henning Kamp said: > In message , Dag-Erling =?iso-8859-1?q?Sm=F8rgrav?= > writes: > >"Poul-Henning Kamp" writes: > >> The thing you overlook is that often when things gets paged out, > >> the system is short on memory and therefore more likely to not do > >> anything productive, whereas when things gets paged in, there are > >> a better chance of some other process being able to use the CPU > >> time productively. If we did predictive pageouts like some of the > >> "serious" mainfram OS's this would be less true. > > > >How hard would it be to get the kernel to write the pages "most > >likely to be swapped out" to swap in the idle loop, to save time if > >/ when they actually need to be swapped out later? > > I don't know :-) > > Quite frankly, given the sizes of RAM we see these days, I think that > paging optimizations may be largely a thing of the past. RAM is like disk space; if it's there, users will consume it. If you have 8GB of RAM, someone will write a program that needs 2gb of RAM, then someone else will decide to run that program in 5 vtys (speaking from experience here). Predictive swapping would be neat. -- Dan Nelson dnelson@allantgroup.com