Received: with ECARTIS (v1.0.0; list linux-xfs); Fri, 06 Dec 2002 09:32:36 -0800 (PST) Received: from zok.sgi.com (zok.SGI.COM [204.94.215.101]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id gB6HWXuR000784 for ; Fri, 6 Dec 2002 09:32:34 -0800 Received: from ledzep.americas.sgi.com (ledzep.americas.sgi.com [192.48.203.134]) by zok.sgi.com (8.12.2/8.12.2/linux-outbound_gateway-1.2) with ESMTP id gB6GeBKp010664 for ; Fri, 6 Dec 2002 08:40:11 -0800 Received: from daisy-e236.americas.sgi.com (daisy-e236.americas.sgi.com [128.162.236.214]) by ledzep.americas.sgi.com (SGI-8.9.3/americas-smart-nospam1.1) with ESMTP id LAA55792; Fri, 6 Dec 2002 11:35:43 -0600 (CST) Received: from rose.americas.sgi.com (rose.americas.sgi.com [128.162.232.98]) by daisy-e236.americas.sgi.com (SGI-8.9.3/SGI-server-1.8) with ESMTP id LAA18724; Fri, 6 Dec 2002 11:35:45 -0600 (CST) Received: from localhost.localdomain (localhost.localdomain [127.0.0.1]) by rose.americas.sgi.com (8.12.5/8.12.5) with ESMTP id gB6HZpIx025713; Fri, 6 Dec 2002 11:35:52 -0600 Subject: RE: Clustered XFS? From: Russell Cattelan To: "N. Richard Solis" Cc: Craig Tierney , linux-xfs@oss.sgi.com In-Reply-To: References: Content-Type: text/plain Content-Transfer-Encoding: 7bit X-Mailer: Ximian Evolution 1.0.8 (1.0.8-10) Date: 06 Dec 2002 11:35:51 -0600 Message-Id: <1039196152.23896.56.camel@rose.americas.sgi.com> Mime-Version: 1.0 X-archive-position: 1995 X-ecartis-version: Ecartis v1.0.0 Sender: linux-xfs-bounce@oss.sgi.com Errors-to: linux-xfs-bounce@oss.sgi.com X-original-sender: cattelan@thebarn.com Precedence: bulk X-list: linux-xfs On Thu, 2002-12-05 at 12:49, N. Richard Solis wrote: > So it's more like DAFS then? No. DAFS is somewhat misnamed actually there are multiple aspect called DDAFS Device (Direct Access FileSystem) UDAFS User and FDAFS File I don't know what progress has been made with UDAFS or FDAFS but DDAFS is simple a protocol to map a remote file (server) to a local device (client) which is really only interesting to things that talk block; like data bases. Note main point of DAFS is to implement a zero copy block transfer mechanism, which isn't to say it can't be used as part of a shared filesystem but DAFS itself isn't intended to provide any of the sharing supported need to do a share file system. > > -----Original Message----- > From: linux-xfs-bounce@oss.sgi.com > [mailto:linux-xfs-bounce@oss.sgi.com]On Behalf Of Craig Tierney > Sent: Thursday, December 05, 2002 1:47 PM > To: N. Richard Solis > Cc: linux-xfs@oss.sgi.com > Subject: Re: Clustered XFS? > > > I have heard the term clustered NFS but I have never seen an actual > implementation. In a clustered NFS solution, is every node actually > talking SCSI (Fibre channel usually) to the disk? That is what CXFS > is doing. Each node physically hits each disk. The metadata server > keep track of what everyone is doing and prevents them from stomping on > each other. > > Craig > > > > > On Thu, Dec 05, 2002 at 12:54:30PM -0500, N. Richard Solis wrote: > > Sounds more like a Clustered NFS solution to me? Would I be on target > with > > that one? > > > > -----Original Message----- > > From: Craig Tierney [mailto:ctierney@hpti.com] > > Sent: Thursday, December 05, 2002 12:41 PM > > To: N. Richard Solis > > Cc: 'Ray Muno '; 'Steve Lord '; linux-xfs@oss.sgi.com > > Subject: Re: Clustered XFS? > > > > > > The client provides the interface so that when a user does a read > > or write, it translates the commands. The server takes care of the layout > > and locking (if supported). Right now you can have a linux box access > > the shared filesystem, but you still need an Irix box to take care of > > the metadata and locking. > > > > I have not worked with CXFS, so I might have exact terminology wrong > > but I think I have the idea right. > > > > Craig > > > > On Thu, Dec 05, 2002 at 12:22:31PM -0500, N. Richard Solis wrote: > > > Client? Server? Please correct me if I'm wrong but should there be a > > > distinction? My understanding of CXFS was that there were enhancements > > made > > > to XFS to cupport concurrent access to a storage pool via a SAN. To do > > > this, you'd need locking/recovery, and some way to ensure that metadata > > > changes were visible to all the servers that had the same filesystem > > > mounted. Where does a client fit into all of this? > > > > > > > > > -----Original Message----- > > > From: linux-xfs-bounce@oss.sgi.com > > > [mailto:linux-xfs-bounce@oss.sgi.com]On Behalf Of Tierney, Craig > > > Sent: Thursday, December 05, 2002 11:59 AM > > > To: 'Ray Muno '; 'Steve Lord ' > > > Cc: 'N. Richard Solis '; 'linux-xfs@oss.sgi.com ' > > > Subject: RE: Clustered XFS? > > > > > > > > > > > > SGI demoed the linux client at SC2002. The server though is still not > > > ready. I think the person I talked to said 'in 6 months' but I could > > > be off there. > > > > > > Craig > > > -----Original Message----- > > > From: Ray Muno > > > To: Steve Lord > > > Cc: N. Richard Solis; linux-xfs@oss.sgi.com > > > Sent: 12/5/02 12:13 PM > > > Subject: Re: Clustered XFS? > > > > > > The promise of a Linux client has been out there as long as CXFS has > > > been > > > around as far as I recall. When is it going to materialize? > > > > > > It has seemed strange to me that SGI can Support Solaris and Windows NT > > > with this environment but Linux support has never been made available. > > > > > > > > > On Thu, Dec 05, 2002 at 10:50:47AM -0600, Steve Lord wrote: > > > > On Thu, 2002-12-05 at 10:43, N. Richard Solis wrote: > > > > > Is anyone working on adding clustering/SAN support to XFS on Linux? > > > > > > > > > > I was looking at the GFS software from Sistina and the OpenGFS > > > project. I'm > > > > > wondering if it's possible to merge the codebases of XFS and OpenGFS > > > to > > > > > provide the clustering functionality on XFS. Am I crazy? > > > > > > > > Go look for CXFS on SGI's home page. There will be something available > > > > for Linux, but it will not be open source, and what parts of it are > > > > available for hardware other than SGI's is under discussion. > > > > > > > > Steve > > > > > > > > -- > > > > > > > > Steve Lord voice: +1-651-683-3511 > > > > Principal Engineer, Filesystem Software email: lord@sgi.com > > > > > > > ======================================================================== > > > ===== > > > > > > Ray Muno > > > http://www.aem.umn.edu/people/staff/muno > > > University of Minnesota e-mail: > > > muno@aem.umn.edu > > > Aerospace Engineering and Mechanics Phone: (612) > > > 625-9531 > > > 110 Union St. S.E. FAX: (612) > > > 626-1558 > > > Minneapolis, Mn 55455 > > > > > > ======================================================================== > > > ===== > > > > > > > > > > > > > > > > > > > > > [[HTML alternate version deleted]] > > > > > > > > > > > > > > > > > > > -- > > Craig Tierney (ctierney@hpti.com) > > > > > > > > -- > Craig Tierney (ctierney@hpti.com) > > >