From owner-freebsd-questions@FreeBSD.ORG Mon Feb 12 15:50:30 2007 Return-Path: X-Original-To: freebsd-questions@freebsd.org Delivered-To: freebsd-questions@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 3C47116A402 for ; Mon, 12 Feb 2007 15:50:30 +0000 (UTC) (envelope-from freebsd-questions-local@be-well.ilk.org) Received: from mail5.sea5.speakeasy.net (mail5.sea5.speakeasy.net [69.17.117.7]) by mx1.freebsd.org (Postfix) with ESMTP id 19DE513C4B4 for ; Mon, 12 Feb 2007 15:50:29 +0000 (UTC) (envelope-from freebsd-questions-local@be-well.ilk.org) Received: (qmail 1056 invoked from network); 12 Feb 2007 15:50:29 -0000 Received: from dsl092-078-145.bos1.dsl.speakeasy.net (HELO be-well.ilk.org) ([66.92.78.145]) (envelope-sender ) by mail5.sea5.speakeasy.net (qmail-ldap-1.03) with SMTP for ; 12 Feb 2007 15:50:29 -0000 Received: by be-well.ilk.org (Postfix, from userid 1147) id 8C98B2842D; Mon, 12 Feb 2007 10:50:28 -0500 (EST) To: Cy Schubert References: <200702121532.l1CFWTE2054040@cwsys.cwsent.com> From: Lowell Gilbert Date: Mon, 12 Feb 2007 10:50:28 -0500 In-Reply-To: <200702121532.l1CFWTE2054040@cwsys.cwsent.com> (Cy Schubert's message of "Mon\, 12 Feb 2007 07\:32\:29 -0800") Message-ID: <44bqjziddn.fsf@be-well.ilk.org> User-Agent: Gnus/5.11 (Gnus v5.11) Emacs/22.0.93 (berkeley-unix) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: freebsd-questions@freebsd.org Subject: Re: Tuning Question X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 12 Feb 2007 15:50:30 -0000 Cy Schubert writes: > Top's output, as is free memory on all O/S's these days, is bogus. It's the > size of the free memory pool which is available for immediate allocation. > Used memory is just as useless. It doesn't matter how much is swapped out, > what matters is how much I/O is being performed to support VM. I know at > work, which is an Oracle ghetto, paging should be kept at a minimum, > especially the SGA. Other apps can afford more. In the case of an average > FreeBSD system it's been guesswork. What I thought you should use top for is tracking the swapping rates. But that's not enough, because the acceptable rates will depend on what else is happening in the system. If the CPU is maxed out anyway (for example), then reducing the swapping will not improve your performance. My general approach to optimization is to identify a problem first. If you can't pin down a performance problem that you want to solve, then you aren't going to be able to prove that any changes fix that problem anyway.