From owner-freebsd-stable Tue Apr 2 10:12:00 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id KAA12573 for stable-outgoing; Tue, 2 Apr 1996 10:12:00 -0800 (PST) Received: from GndRsh.aac.dev.com (GndRsh.aac.dev.com [198.145.92.241]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id KAA12565 for ; Tue, 2 Apr 1996 10:11:57 -0800 (PST) Received: (from rgrimes@localhost) by GndRsh.aac.dev.com (8.6.12/8.6.12) id KAA03701; Tue, 2 Apr 1996 10:11:35 -0800 From: "Rodney W. Grimes" Message-Id: <199604021811.KAA03701@GndRsh.aac.dev.com> Subject: Re: SCSI Performance under -stable/-current To: msmith@atrad.adelaide.edu.au (Michael Smith) Date: Tue, 2 Apr 1996 10:11:35 -0800 (PST) Cc: Zeki@Dilbert.Oasysinc.com, stable@FreeBSD.ORG In-Reply-To: <199604021022.TAA14221@genesis.atrad.adelaide.edu.au> from Michael Smith at "Apr 2, 96 07:52:07 pm" X-Mailer: ELM [version 2.4ME+ PL11 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > Zeki Basbuyuk stands accused of saying: > > > How do you get this speed? I thought there is a physical limitation > > on how much speed you can get for a hard drive. The calculation is > > as follows. 120 rps (revs per sec assuming 7200 rpm drive) X 40 > > sectors per track (average or above average) X 512 bps (bytes per > > sector) = 2.4 MB /sec. To get 6 MB either spt needs to increase to > > 100 or so or rotation needs to be increased. The flaw in the above is: [From www.quantum.com Fireball 640/1280 pages] Sectors per Track 95 to 177 Rotational Speed (RPM) 5,400 90 * 177 * 512 == 8.156MB/s, 90 * 95 * 512 == 4.377MB/s, your 40 spt is 5 year old technology, current technology is pushing this way up, the next generation drives are going to be on the order of 200+ spt. > > One should always temper theory with reality. You can go a lot faster by > striping your sectors across several surfaces, for example. Very very very rare. > Reality shows us that the more agressive drives will happily outrun the > sustained rate your envelope calculation suggests, so I guess reality is > more complicated that you're allowing for 8) He just used bad values for the drive. One other important number to pull out of the data sheet is maximum head to drive electronics bit rates, which many vendors don't specify in there data sheets. -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD