From owner-freebsd-questions@FreeBSD.ORG Wed May 6 06:21:22 2009 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 36E3F106567A for ; Wed, 6 May 2009 06:21:22 +0000 (UTC) (envelope-from wojtek@wojtek.tensor.gdynia.pl) Received: from wojtek.tensor.gdynia.pl (wojtek.tensor.gdynia.pl [IPv6:2001:4070:101:2::1]) by mx1.freebsd.org (Postfix) with ESMTP id 4C8388FC15 for ; Wed, 6 May 2009 06:21:19 +0000 (UTC) (envelope-from wojtek@wojtek.tensor.gdynia.pl) Received: from wojtek.tensor.gdynia.pl (localhost [IPv6:::1]) by wojtek.tensor.gdynia.pl (8.14.3/8.14.3) with ESMTP id n466KRZB028372; Wed, 6 May 2009 08:20:27 +0200 (CEST) (envelope-from wojtek@wojtek.tensor.gdynia.pl) Received: from localhost (wojtek@localhost) by wojtek.tensor.gdynia.pl (8.14.3/8.14.3/Submit) with ESMTP id n466KPiY028369; Wed, 6 May 2009 08:20:27 +0200 (CEST) (envelope-from wojtek@wojtek.tensor.gdynia.pl) Date: Wed, 6 May 2009 08:20:25 +0200 (CEST) From: Wojciech Puchar To: Yuri In-Reply-To: <4A00B3FA.5050905@rawbw.com> Message-ID: References: <4A00B3FA.5050905@rawbw.com> User-Agent: Alpine 2.00 (BSF 1167 2008-08-23) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Cc: freebsd-questions@freebsd.org Subject: Re: What is the highest hard drive read/write speed you were able to achieve by entire disk mirroring or striping? 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: Wed, 06 May 2009 06:21:22 -0000 > I am seeing 85MB/s as a speed of a single Hitachi 1TB HD. > How high can you go by mirroring or striping 2, 3, 4 harddrives? mirroring - the same, just with 2 processes reading both can get the bandwidth. make sure you use -s high enough (like 1048576) doing gmirror label stripping - the same, or 2,3,4 times, depends how you configure. for highest transfer and lowest concurrency (you mostly read huge files with one process) - use small stripe size. for lowest transfer (=1 disk) and highest concurency - use very huge stripe size like 512MB, so simply different process reading different things can hit different drives, but each I/O isn't spread.