From owner-freebsd-stable@FreeBSD.ORG Wed Jan 14 05:42:42 2004 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id DDF3416A4D2 for ; Wed, 14 Jan 2004 05:42:42 -0800 (PST) Received: from mx0.dmpriest.net.uk (mx0.dmpriest.net.uk [62.13.128.30]) by mx1.FreeBSD.org (Postfix) with ESMTP id E950043D31 for ; Wed, 14 Jan 2004 05:42:39 -0800 (PST) (envelope-from kpielorz@tdx.co.uk) Received: from rainbow (adsl-blk-62-13-130-225.dmpriest.net.uk [62.13.130.232] (may be forged)) by mx0.dmpriest.net.uk (8.11.6/8.11.6/Kp) with ESMTP id i0EDdlb15899; Wed, 14 Jan 2004 13:39:47 GMT Date: Wed, 14 Jan 2004 13:42:40 +0000 From: Karl Pielorz To: Ari Suutari , freebsd-stable@freebsd.org Message-ID: <742320171.1074087760@rainbow> In-Reply-To: <200401141527.11930.ari@suutari.iki.fi> References: <200401141453.50150.ari@suutari.iki.fi> <741002750.1074086443@rainbow> <200401141527.11930.ari@suutari.iki.fi> X-Mailer: Mulberry/3.1.0 (Win32) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline Subject: Re: Adaptect raid performance with FreeBSD X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 14 Jan 2004 13:42:43 -0000 --On 14 January 2004 15:27 +0200 Ari Suutari wrote: > Hi, > > On Wednesday 14 January 2004 15:20, Karl Pielorz wrote: >> > So, I get only about 25MB/s. Shouldn't I be getting something >> > like 70 MB/s, or even more since there are two disks that >> > can server read requests ? >> >> Hmmm, what happens if you run two of those at once? > > Running 5 dd's at once gives in iostat: > > [snip] > > The peak seems to be 39 Mb/s. Was that the peak 'each' - or combined? - i.e. what was the combined throughput for 2 dd's running? >> PCI latency.. >> >> But, I can't remember seeing adjusting it making any real-world >> difference - what happens if you max it out (e.g. 128, or 256)? > > I'll have to try this. However, I also doubt that it is not going > to change anything. Wouldn't be the first time :) Give two dd's a run at the same, with 64k blocksizes and see what you get for each one... We have a machine at the office with the same controller in it, and some IBM 36Gb 10K drives, but unfortunately it's not in any fit state to dd anything at the moment, else I'd give it a while :) -Karl