From owner-freebsd-hackers Thu Jun 19 19:09:48 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id TAA03738 for hackers-outgoing; Thu, 19 Jun 1997 19:09:48 -0700 (PDT) Received: from sendero.i-connect.net (sendero-ppp.i-Connect.Net [206.190.143.100]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id TAA03728; Thu, 19 Jun 1997 19:09:39 -0700 (PDT) Received: (from shimon@localhost) by sendero.i-connect.net (8.8.5/8.8.5) id TAA01977; Thu, 19 Jun 1997 19:09:38 -0700 (PDT) Message-ID: X-Mailer: XFMail 1.2-alpha [p0] on FreeBSD Content-Type: text/plain; charset=iso-8859-8 Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: Date: Thu, 19 Jun 1997 19:09:38 -0700 (PDT) Organization: Atlas Telecom From: Simon Shapiro To: Brian Tao Subject: Re: Announcement: New DPT RAID Controller Driver Available Cc: FREEBSD-SCSI , FREEBSD-HACKERS Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi Brian Tao; On 18-Jun-97 you wrote: > On Mon, 16 Jun 1997, Simon Shapiro wrote: > > > > I have evaluated, tested and even participated in the design of more > > than one I/O subsystem in my short life. The StorageWorks solution > > is the best I have ever seen. not perfect, Just the BEST. > > I should add that NetApp also uses the DEC StorageWorks shelves > and canisters for their 2GB and 4GB configurations, and if it's good > enough for them, it's good enough for me. ;-) Mounting drives > inside the canisters is a little tricky at first, but you get the hang > of it pretty quickly. The ones we have are plastic, so I imagine the > heat dissipation isn't the best. Netapp went with a Eurologics > product (metal extruded cases) for their 9GB shelves. You know, the canister problem is an interesting one. I was once involved in a project where we build a database server with 3,000 (no typo) drives, so canisters and packing density were somewhat important :-). An excellent mechanical engineer designed a wonderful metal system (both canisters and carriers). We had up to 30% perofrmance loss on random seeks with system. It baffled anyone until another engineer decided to test the drives on his desk, outside the box, outside the carriers - Yes, the performance was back. Turns out soft errors were masked and the rigid but ringing-resonating nature of steel and the high packing density caused drives to resonate the cabinets and cause miseeks. For other, unrelated problems, we decided to use the DEC canisters. Smug as a cat with a bird in mouth we loaded 200 drives into a cabinet, smirking and making snide remarks about the silly plastic carriers - No data loss, no perfromance loss. Asking the DEC engineers about why they choose platic, they just smiled and shrugged... Simon BTW, the little flex-circuit between the drive and the canister plug makes you think too. Simon