Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751371AbVJFVM5 (ORCPT ); Thu, 6 Oct 2005 17:12:57 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751375AbVJFVM5 (ORCPT ); Thu, 6 Oct 2005 17:12:57 -0400 Received: from mail.tmr.com ([64.65.253.246]:50592 "EHLO gaimboi.tmr.com") by vger.kernel.org with ESMTP id S1751371AbVJFVM4 (ORCPT ); Thu, 6 Oct 2005 17:12:56 -0400 Message-ID: <4345940C.1080307@tmr.com> Date: Thu, 06 Oct 2005 17:15:56 -0400 From: Bill Davidsen Organization: TMR Associates Inc, Schenectady NY User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.11) Gecko/20050729 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Guy CC: "'Holger Kiehl'" , "'Mark Hahn'" , "'linux-raid'" , "'linux-kernel'" Subject: Re: Where is the performance bottleneck? References: <200509300452.j8U4qKw17804@www.watkins-home.com> In-Reply-To: <200509300452.j8U4qKw17804@www.watkins-home.com> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1935 Lines: 61 Guy wrote: > > >>-----Original Message----- >>From: linux-raid-owner@vger.kernel.org [mailto:linux-raid- >>owner@vger.kernel.org] On Behalf Of Bill Davidsen >>Sent: Wednesday, September 28, 2005 4:05 PM >>To: Guy >>Cc: 'Holger Kiehl'; 'Mark Hahn'; 'linux-raid'; 'linux-kernel' >>Subject: Re: Where is the performance bottleneck? >> >>Guy wrote: >> >> >> >>>In most of your results, your CPU usage is very high. Once you get to >>> >>> >>about >> >> >>>90% usage, you really can't do much else, unless you can improve the CPU >>>usage. >>> >>> >>> >>That seems one of the problems with software RAID, the calculations are >>done in the CPU and not dedicated hardware. As you move to the top end >>drive hardware the CPU gets to be a limit. I don't remember off the top >>of my head how threaded this code is, and if more CPUs will help. >> >> > >My old 500MHz P3 can xor at 1GB/sec. I don't think the RAID5 logic is the >issue! Also, I have not seen hardware that fast! Or even half as fast. >But I must admit, I have not seen a hardware RAID5 in a few years. :( > > 8regs : 918.000 MB/sec > 32regs : 469.600 MB/sec > pIII_sse : 994.800 MB/sec > pII_mmx : 1102.400 MB/sec > p5_mmx : 1152.800 MB/sec >raid5: using function: pIII_sse (994.800 MB/sec) > >Humm.. It did not select the fastest? > Maybe. There was discussion on this previously, but the decision was made to us sse when available because it is nicer to cache, or uses fewer registers, or similar. In any case fewer undesirable side effects. -- bill davidsen CTO TMR Associates, Inc Doing interesting things with small computers since 1979 - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/