[pnfs] Linux pNFS road map

[Muntz, Daniel]

Yes, there's a bunch of stuff happening in userspace in the current
implementation.  Some (most, all?) will move into the kernel.  The
existing userspace design was done primarily to get something out
quickly, which we could also modify/debug easily.  There's pleanty of
stuff on the todo list, which we'll put out with the code RSN.
 
  -Dan

________________________________

From: William A. (Andy) Adamson [mailto:andros at citi.umich.edu] 
Sent: Tuesday, December 11, 2007 8:23 AM
To: J. Bruce Fields
Cc: Muntz, Daniel; pnfs at linux-nfs.org; peter honeyman
Subject: Re: [pnfs] Linux pNFS road map




On Dec 11, 2007 9:41 AM, J. Bruce Fields <bfields at fieldses.org> wrote:


	On Mon, Dec 10, 2007 at 12:25:42PM -0800, Muntz, Daniel wrote:
	
	> This addresses the client side.  Trond has suggested that the
client
	> code won't go into mainline without an open source server
queued up 
	> within an rc or two of the client (or vice versa).  We've been
working
	> on a simple pNFS server (spNFS) that might fill this role, and
we're
	> close to putting it up on linux-nfs.org.  I'll attach the
README file
	> (incomplete) I've been working on for the release.
	
	
	By the time you're done implementing I/O to the metadata server
and all
	metadata operations (create, rename, remove,...), doesn't spNFS 
	essentially end up being a complete filesystem in userspace?


there are many design decisions yet to be made.....

-->Andy



	
	--b.
	
	_______________________________________________
	pNFS mailing list
	pNFS at linux-nfs.org
	http://linux-nfs.org/cgi-bin/mailman/listinfo/pnfs
	
	


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://linux-nfs.org/pipermail/pnfs/attachments/20071211/32d0d63a/attachment.htm