From owner-freebsd-questions Thu Jul 11 08:18:29 1996 Return-Path: owner-questions Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id IAA22114 for questions-outgoing; Thu, 11 Jul 1996 08:18:29 -0700 (PDT) Received: from iworks.InterWorks.org (deischen@iworks.interworks.org [128.255.18.10]) by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id IAA22097 for ; Thu, 11 Jul 1996 08:18:22 -0700 (PDT) Received: by iworks.InterWorks.org (1.37.109.8/16.2) id AA12246; Thu, 11 Jul 1996 10:15:28 -0500 Message-Id: <9607111515.AA12246@iworks.InterWorks.org> Date: Thu, 11 Jul 1996 10:15:28 -0500 From: "Daniel M. Eischen" To: mark@ucsalf.ac.uk Subject: Re: Does Adaptec 7880 Ultra support simultaneous wide and narrow transfers? Cc: freebsd-questions@freebsd.org Sender: owner-questions@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > On Thu, 11 Jul 1996, Daniel M. Eischen wrote: > > When did you get your -stable system? There were some bug fixes that were incorporated > > into the aic7xxx driver in late June. Make sure you're running a -stable after that > > date. > > Cheers, that seems to have fixed it. ~900K/s now. Still I thought two > 7200rpm drives would be at least in the M/s range? Hmm. Here's a Bonnie of a sytem we have here at work. It's a P166, 32M RAM, 2940UW, and a couple of 2.1GB Seagate Ultra Wide drives at 5400RPM: -------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 3543 62.1 3584 14.2 1404 11.0 2515 97.6 3976 15.4 60.6 2.5 Tagged queueing and SCB paging are enabled. With simultaneous bonnies run on each disk: -------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 2823 49.7 3478 14.7 1389 11.2 1795 72.2 3972 15.6 57.6 2.6 100 2757 51.0 4607 20.7 1436 11.6 3329 54.2 5174 21.2 65.9 3.1 Seems like you should be getting more than 900K/s... > Okay, done both. Seems okay. Drive seems to make a different noise on > boot-up now. Different seek patterns with the new code or just pure > imagination? Yeah, with tagged queueing enabled, the device(s) can process commands in any order it wants to increase performance (less seeking). Plus, you're sending more commands to the device and making it work harder. Without tagged queueing enabled, you limit the device to processing one command at a time. Tagged queueing is your friend :-) Dan Eischen deischen@iworks.InterWorks.org