From owner-freebsd-questions@FreeBSD.ORG Thu Feb 26 14:08:45 2009 Return-Path: Delivered-To: questions@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id F3060106567F for ; Thu, 26 Feb 2009 14:08:44 +0000 (UTC) (envelope-from josh@tcbug.org) Received: from out1.smtp.messagingengine.com (out1.smtp.messagingengine.com [66.111.4.25]) by mx1.freebsd.org (Postfix) with ESMTP id BA7A58FC2B for ; Thu, 26 Feb 2009 14:08:44 +0000 (UTC) (envelope-from josh@tcbug.org) Received: from compute2.internal (compute2.internal [10.202.2.42]) by out1.messagingengine.com (Postfix) with ESMTP id 233252A4D38; Thu, 26 Feb 2009 09:08:44 -0500 (EST) Received: from heartbeat2.messagingengine.com ([10.202.2.161]) by compute2.internal (MEProxy); Thu, 26 Feb 2009 09:08:44 -0500 X-Sasl-enc: Zm8Y6PfXnbrIwXwAzXbH5qOAYvQQXL2ef8Jcy+MVV2hH 1235657323 Received: from [10.0.1.198] (c-66-41-132-62.hsd1.mn.comcast.net [66.41.132.62]) by mail.messagingengine.com (Postfix) with ESMTPSA id 9982D3556D; Thu, 26 Feb 2009 09:08:43 -0500 (EST) Message-Id: <521399F6-5C55-4ECD-8DC7-46DC7432F250@tcbug.org> From: Josh Paetzel To: Wojciech Puchar In-Reply-To: Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes Content-Transfer-Encoding: 7bit Mime-Version: 1.0 (Apple Message framework v930.3) Date: Thu, 26 Feb 2009 08:08:42 -0600 References: X-Mailer: Apple Mail (2.930.3) Cc: questions@freebsd.org Subject: Re: SAS drives seem slow 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: Thu, 26 Feb 2009 14:08:54 -0000 On Feb 26, 2009, at 12:28 AM, Wojciech Puchar wrote: >> FreeBSD services.tcbug.org 8.0-CURRENT FreeBSD 8.0-CURRENT #0: Mon >> Feb 16 21:07:14 UTC 2009 root@:/usr/obj/usr/src/sys/SERVICES >> amd64 >> >> But 6.x and 7.x give similar results. >> >> The disks themselves are capable of sequential read/write in the >> 180 Meg/sec range, so I'm trying to understand why I'm being told >> they are 100 Meg/sec, and why that seems to be their real world >> performance cap. >> > > probably you use RAID5 Hrmm, I should have included drive configuration. root@services /home/jpaetzel ->tw_cli /c0 show Unit UnitType Status %RCmpl %V/I/M Stripe Size(GB) Cache AVrfy ------------------------------------------------------------------------------ u0 RAID-1 OK - - - 135.031 ON ON u1 RAID-1 OK - - - 298.013 ON ON VPort Status Unit Size Type Phy Encl-Slot Model ------------------------------------------------------------------------------ p0 OK u0 136.98 GB SAS 0 - FUJITSU MBA3147RC p1 OK u0 136.98 GB SAS 1 - FUJITSU MBA3147RC p2 OK u1 298.09 GB SATA 2 - WDC WD3200AAKS-00SB p3 OK u1 298.09 GB SATA 3 - WDC WD3200AAKS-00SB But I get similar results when I connect just a single SAS drive and export it as a raw device. It's also worth noting that the SATA drive array is reported as 100MB/sec transfers, even though the drives aren't capable of anything close to that, unless they are reading from cache, in which case SATA2 is capable of more like 300MB/sec...but then so is SAS... Thanks, Josh Paetzel