From owner-freebsd-alpha Tue Aug 24 12:56:41 1999 Delivered-To: freebsd-alpha@freebsd.org Received: from smtp03.primenet.com (smtp03.primenet.com [206.165.6.133]) by hub.freebsd.org (Postfix) with ESMTP id 17FDE158EC for ; Tue, 24 Aug 1999 12:55:58 -0700 (PDT) (envelope-from tlambert@usr09.primenet.com) Received: (from daemon@localhost) by smtp03.primenet.com (8.9.3/8.9.3) id MAA17675; Tue, 24 Aug 1999 12:55:39 -0700 (MST) Received: from usr09.primenet.com(206.165.6.209) via SMTP by smtp03.primenet.com, id smtpdAAA3vaGxI; Tue Aug 24 12:55:29 1999 Received: (from tlambert@localhost) by usr09.primenet.com (8.8.5/8.8.5) id MAA01709; Tue, 24 Aug 1999 12:55:36 -0700 (MST) From: Terry Lambert Message-Id: <199908241955.MAA01709@usr09.primenet.com> Subject: Re: Clustering? To: dg@root.com Date: Tue, 24 Aug 1999 19:55:36 +0000 (GMT) Cc: mikel@zso.dec.com, freebsd-alpha@FreeBSD.ORG In-Reply-To: <199908241845.LAA15425@implode.root.com> from "David Greenman" at Aug 24, 99 11:45:25 am X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > >These pairs are tied together with differential SCSI bus to shared storage > >for "clustering". Specifically, they were used for NT "Wolfpack". > > > >Does anyone know if FreeBSD already has the smarts to handle this sort of > >configuration, or would I be having write device drivers to make use of > >it? You might have to write the drivers. If you can get hardware docs along with the machines that would be a good thing to do, in any case. > >Can you see any reason I should not try to acquire this equipment? I > >could skip the shared storage, differential SCSI, etc. and save some money > >if it's known the FreeBSD would not like it. > > > >Please reply to me directly ('though I am on the list) as the rest of the > >list probably won't be interested. > > FreeBSD doesn't support shared filesystems. You would need to add support > for that into FFS and presumably get or implement a distributed lock manager. The David Sarnoff stuff supports this. It would not be too difficult to hack up the disctubted memory coherency manager to use low level SCSI I/O primitives to support this. Also, SGI continues to threaten to release XFS sources, which are capable of supporting this kind of configuration (the biggest issue there is the same as on NT, since NTFS has a single log lock per volume, and XFS has a single journal lock per volume). Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message