From owner-freebsd-hardware@FreeBSD.ORG Wed Mar 3 20:21:55 2004 Return-Path: Delivered-To: freebsd-hardware@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id D0C0916A4CF for ; Wed, 3 Mar 2004 20:21:55 -0800 (PST) Received: from salvador.pacific.net.sg (salvador.pacific.net.sg [203.120.90.219]) by mx1.FreeBSD.org (Postfix) with SMTP id D5C3643D2F for ; Wed, 3 Mar 2004 20:21:54 -0800 (PST) (envelope-from oceanare@pacific.net.sg) Received: (qmail 735 invoked from network); 4 Mar 2004 04:21:53 -0000 Received: from unknown (HELO maxwell6.pacific.net.sg) (203.120.90.212) by with SMTP; 4 Mar 2004 04:21:52 -0000 Received: from pacific.net.sg ([210.24.202.170]) by maxwell6.pacific.net.sg with ESMTP id <20040304042152.OGSN9972.maxwell6.pacific.net.sg@pacific.net.sg>; Thu, 4 Mar 2004 12:21:52 +0800 Message-ID: <4046AEE0.4010201@pacific.net.sg> Date: Thu, 04 Mar 2004 12:21:52 +0800 From: Erich Dollansky Organization: oceanare pte ltd User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.6) Gecko/20040113 X-Accept-Language: en, en-us, de MIME-Version: 1.0 To: Simon References: <20040303233733.9402F43D1D@mx1.FreeBSD.org> In-Reply-To: <20040303233733.9402F43D1D@mx1.FreeBSD.org> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit cc: "hardware@freebsd.org" Subject: Re: Xeon w/ L3 1MB cache vs Xeon w/o L3 cache X-BeenThere: freebsd-hardware@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: General discussion of FreeBSD hardware List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 04 Mar 2004 04:21:55 -0000 Hi, Simon wrote: > Has anyone done any comparison to see if extra L3 cache on Xeon CPUs > provides any benefit to FreeBSD's kernel/core services and various user > services' (http/email/ftp/databases) performance? I read that L3 can make > things slower instead of faster in cases where L3 is not utilized and the CPU > is forced to access it anyway (when L1/L2 don't hold the data it wants), > because it exists. Would anyone with personal experience with Xeon CPUs > with and without L3 cache comment on the impact of L3 cache. Any links > to any related articles would be appreciated, as well. > I only have experience with other CPUs with L3 cache. L3 keeps them faster under high load conditions. It made a small negative impact under low load conditions because their L3 cache was much slower than their L1 and L2 caches. If you have the high load it is worth the high price they cost. I do not have any links for this. The work was done some time ago in a typical ISP environment. The tests included only FTP and database. Erich