From owner-freebsd-performance@FreeBSD.ORG Fri May 20 03:13:34 2005 Return-Path: Delivered-To: freebsd-performance@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 2B85B16A4CE for ; Fri, 20 May 2005 03:13:34 +0000 (GMT) Received: from virgin.ww.net (virgin.kolo.net [216.218.215.99]) by mx1.FreeBSD.org (Postfix) with ESMTP id D361E43D76 for ; Fri, 20 May 2005 03:13:33 +0000 (GMT) (envelope-from alexis@virgin.ww.net) Received: from virgin.ww.net (localhost [127.0.0.1]) by virgin.ww.net (8.13.1/8.13.1) with ESMTP id j4K3DWtq098688 for ; Fri, 20 May 2005 05:13:32 +0200 (CEST) (envelope-from alexis@virgin.ww.net) Received: (from alexis@localhost) by virgin.ww.net (8.13.1/8.13.1/Submit) id j4K3DWCV098687 for freebsd-performance@freebsd.org; Fri, 20 May 2005 05:13:32 +0200 (CEST) (envelope-from alexis) Date: Fri, 20 May 2005 05:13:32 +0200 From: Alexis Yushin To: freebsd-performance@freebsd.org Message-ID: <20050520031332.GA98548@virgin.ww.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.4.2.1i X-Spam-Status: No, score=-2.8 required=5.0 tests=ALL_TRUSTED autolearn=failed version=3.0.2 X-Spam-Checker-Version: SpamAssassin 3.0.2 (2004-11-16) on virgin.ww.net Subject: Bad 3ware 9000 performance 5.4? X-BeenThere: freebsd-performance@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Performance/tuning List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 20 May 2005 03:13:34 -0000 Hi, I've been browsing through the mailing list archive, but could not find the answer/conclusion/solution to this... Running FreeBSD 5.4... three machines with twe/3ware 8000 and three twa/ 3ware 9500-8 controllers. All of them running RAID 1+0 without write cache since we dont have battery backup unit. All of them Seagate 7200 SATA. I get about 46Mb/s write on the 8000 series and only 5(!)MB/s on the 9500 series. How is this possible? Machines are in production so I can do very little testing... Any ideas? Alexis