From owner-aic7xxx Fri Aug 14 15:54:26 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id PAA07315 for aic7xxx-outgoing; Fri, 14 Aug 1998 15:54:26 -0700 (PDT) (envelope-from owner-aic7xxx@FreeBSD.ORG) Received: from feral.com (gw100.feral.com [192.67.166.129]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id PAA07308 for ; Fri, 14 Aug 1998 15:54:21 -0700 (PDT) (envelope-from mjacob@feral.com) Received: from localhost (mjacob@localhost) by feral.com (8.8.6/8.8.6) with SMTP id PAA27400; Fri, 14 Aug 1998 15:53:34 -0700 Date: Fri, 14 Aug 1998 15:53:34 -0700 (PDT) From: Matthew Jacob X-Sender: mjacob@feral-gw To: "Justin T. Gibbs" cc: Ross Harvey , matti.aarnio@sonera.fi, aic7xxx@FreeBSD.ORG Subject: Re: AHA2790UW has speed-limit problems ? In-Reply-To: <199808142240.QAA05851@pluto.plutotech.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-aic7xxx@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Sure- but a single-initiator/single-target system is handled well w/o even going to Ultra right now. I suppose LVD handles a longer wire, but with current cabling w/o Ultra you're good for 5 meters or so. It's the case where you want 10-20 drives per 'initiator' and 4-8 'initiators' per system where it gets very useful. And I look at from a system integrators point of view- the cost of a SE/LVD card from adaptec may be the same to me as a FC card. On Fri, 14 Aug 1998, Justin T. Gibbs wrote: > > > >It's exactly this kind of discussion that has encouraged me to think > >that Fibre Channel is a better technology when you're getting to these > >speeds. The cost differential between this and parallel SCSI is getting > >to be pretty small. > > I wish IBM hadn't botched the marketing of SSA. It's so much nicer than > Fiber Channel. The concern we have here at Pluto about Fiber Channel is > the power cost. We couldn't put 20 fiber channel drives into our system > without bumping up to much more expensive, larger capacity, power supplies. > There is also the fault tolerance issue and expense. We're single target/ > single initiator so we can "play to air" even if a drive hangs on the wire. > With SE/LVD SCSI, we can do this for ~$20 per SCSI HA chip. SSA would > have solved this problem with it's dual port nature without forcing us to a > single controller per drive configuration, but with fiber channel, you have > no choice. I can't imagine being able to build a cost effective single > initiator/single target system with fiber channel. The tranceiver cost > alone would kill us. > > -- > Justin > > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe aic7xxx" in the body of the message