From owner-freebsd-current Fri Feb 19 13:17:21 1999 Delivered-To: freebsd-current@freebsd.org Received: from panzer.plutotech.com (panzer.plutotech.com [206.168.67.125]) by hub.freebsd.org (Postfix) with ESMTP id 9F38B11923 for ; Fri, 19 Feb 1999 13:17:18 -0800 (PST) (envelope-from ken@panzer.plutotech.com) Received: (from ken@localhost) by panzer.plutotech.com (8.9.2/8.8.5) id OAA04024; Fri, 19 Feb 1999 14:15:52 -0700 (MST) From: "Kenneth D. Merry" Message-Id: <199902192115.OAA04024@panzer.plutotech.com> Subject: Re: Slow seq. write on Seagate ST36530N In-Reply-To: <199902182224.XAA00605@trantor.xs4all.nl> from Paul van der Zwan at "Feb 18, 1999 11:24:21 pm" To: paulz@trantor.xs4all.nl (Paul van der Zwan) Date: Fri, 19 Feb 1999 14:15:52 -0700 (MST) Cc: freebsd-current@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL43 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Paul van der Zwan wrote... > > I am having some performance problems on my -current ( update last weekend) > I hooked up a new Seagate ST36530N yesterday ( connected to an Adaptec 2940U) > and sequential write is very slow. > Compared to an IBM DORS-32160 connected to the same controller ( even the same > cable) it is half as fast. > Iozone auto shows the following : > > Seagate > IOZONE: Performance Test of Sequential File I/O -- V2.01 (10/21/94) > By Bill Norcott > > Operating System: FreeBSD 2.x -- using fsync() > > IOZONE: auto-test mode > > MB reclen bytes/sec written bytes/sec read > 1 512 5835553 22369621 > 1 1024 3627506 33554432 > 1 2048 3441480 44739242 > 1 4096 4329604 44739242 > 1 8192 3121342 67108864 [ ... ] > IBM > MB reclen bytes/sec written bytes/sec read > 1 512 3728270 22369621 > 1 1024 4067203 26843545 > 1 2048 3947580 67108864 > 1 4096 3728270 44739242 > 1 8192 3834792 67108864 > 2 512 4549753 13421772 > 2 1024 4194304 44739242 > 2 2048 3890368 53687091 [ ... ] > Bonnie shows the following: > Seagate > -------Sequential Output-------- ---Sequential Input-- --Random-- > -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks--- > Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU > 100 3251 44.0 1307 4.0 2285 11.5 5006 69.0 8644 23.0 115.1 4.2 > IBM > 100 3333 45.0 2533 8.8 1878 10.1 4244 58.2 5140 19.7 76.4 3.3 > > If I interpret it correctly the Seagate is faster in everything but sequential > writes. > dmesg shows the following : > > da1 at ahc0 bus 0 target 1 lun 0 > da1: Fixed Direct Access SCSI-2 device > da1: 10.0MB/s transfers (10.0MHz, offset 15), Tagged Queueing Enabled > da1: 6208MB (12715920 512 byte sectors: 255H 63S/T 791C) > da0 at ahc0 bus 0 target 0 lun 0 > da0: Fixed Direct Access SCSI-2 device > da0: 10.0MB/s transfers (10.0MHz, offset 15), Tagged Queueing Enabled > da0: 2063MB (4226725 512 byte sectors: 255H 63S/T 263C) > > Anybody an idea ?? Andreas Klemm has had similar trouble, as he pointed out. Can you check and see whether or not you have write caching turned on for your disk? I have seen problems with sequential writes that appear to be caused by conflicts between FreeBSD's caching policy and disk caching policies. These problems often go away when you disable write caching on a disk. The Write Cache Enable (WCE) bit is in mode page 8. To check it: camcontrol modepage -n da -u 1 -v -m 8 To edit the mode page: camcontrol modepage -n da -u 1 -v -m 8 -e Let me know whether that affects the problem at all. Ken -- Kenneth Merry ken@plutotech.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message