From owner-freebsd-questions Tue Nov 3 14:29:58 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id OAA24023 for freebsd-questions-outgoing; Tue, 3 Nov 1998 14:29:58 -0800 (PST) (envelope-from owner-freebsd-questions@FreeBSD.ORG) Received: from chopin.seattleu.edu (chopin.seattleu.edu [206.81.198.70]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id OAA24013 for ; Tue, 3 Nov 1998 14:29:55 -0800 (PST) (envelope-from hodeleri@seattleu.edu) Received: from seattleu.edu ([172.17.25.95]) by chopin.seattleu.edu (8.8.5/8.8.5) with ESMTP id OAA08488; Tue, 3 Nov 1998 14:29:31 -0800 (PST) Message-ID: <363F837B.2F4BF749@seattleu.edu> Date: Tue, 03 Nov 1998 14:28:11 -0800 From: Eric Hodel X-Mailer: Mozilla 4.5 [en] (Win95; I) X-Accept-Language: en MIME-Version: 1.0 To: Graeme Tait , freebsd-questions@FreeBSD.ORG Subject: Re: File system performance References: <363F7AA3.22254A9C@echidna.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Stabbing in the dark, but the interleave on the SCSI drive may not be optimized for the small 1k files you use. To do this you would have to do a low-level format of you SCSI drive, then reinstall your software. Read the manual or contact Seagate to see if a low-level format is allowed for your drive, I've heard that it can trash some types of drives. Graeme Tait wrote: > > I have a situation that involves manipulating large numbers of small > files of about 1k each. I recently noticed a strange performance > comparison between my "play" system (a 486-DX2/66/16MB with > run-of-the-mill IDE drives and a Promise caching controller, running > 2.2.6R) and "production" system (Pentium-II/400/256MB with Ultra 2 LVD > SCSI and 4.5GB Seagate Cheetah drives, running 2.2.7S/CAM). > > When deleting these files (rm -rf), the 486 does it with a minimum of > fuss (no trashing of the disk heads) at about 120 files a second. Disk > operations occur ever second or so with intervals between. > > The production system only manages about 40 files a second, and you can > hear that the disk heads are going for their life, vibrating back and > forth between tracks. It's impressive to hear the speed of the head > positioning, but the end result is hardly so. > > In both cases, the files being deleted were expanded from the same > archive, into a relatively empty filesystem (built with the default > fragment size, etc.). > > Similar behaviour occurs when the gzipped archive is expanded, but in > this case the systems are about equal in speed. > > In other filesystem comparisions, with larger files, the production > system runs rings around the test system, as you would expect. > > What is happening here? Is maybe something configured wrong on the > production system? > > -- > Graeme Tait - Echidna > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-questions" in the body of the message To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message