From owner-freebsd-hackers Sat Mar 23 19:46:19 1996 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id TAA23687 for hackers-outgoing; Sat, 23 Mar 1996 19:46:19 -0800 (PST) Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id TAA23677 for ; Sat, 23 Mar 1996 19:46:14 -0800 (PST) Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.12/8.6.9) id OAA26704; Sun, 24 Mar 1996 14:42:14 +1100 Date: Sun, 24 Mar 1996 14:42:14 +1100 From: Bruce Evans Message-Id: <199603240342.OAA26704@godzilla.zeta.org.au> To: hackers@freebsd.org, toor@dyson.iquest.net Subject: Re: Interesting IDE perf results Cc: davidg@root.com Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >I agree that SCSI is generally better for high-end solutions, but I >was somewhat pleasantly suprised to see these results with a 1.6GB WD Caviar >IDE drive... IDE should be slightly faster provided the (total) PIO transfer speed is at least as fast as the (total) SCSI transfer speed and the drives are the same speed. The transfer speeds usually favor IDE for _one_ modern IDE controller/bus (15MB/sec) and _one_ not so modern SCSI controller (10MB/ sec). >... >File './Bonnie.212', size: 104857600 >Seeker 1...Seeker 2...Seeker 3...start 'em...done...done...done... > -------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 3837 70.6 5260 24.7 1782 9.6 4600 68.1 5482 17.4 92.7 3.8 Bonnie makes IDE look much better than it is by not counting interrupt overhead in %CPU. It only makes busmastering SCSI look better than it is by not counting bus overhead in %CPU :-). Bruce