Murray Altheim <m.altheim@open.ac.uk> writes: (01)
% > Please talk more about this NODAL idea. (02)
% Have you read Lee Iverson's paper on NODAL? It's pretty explicit about
% the contents and structure: (03)
% http://nodal.sourceforge.net/NODAL-WhitePaper.html
% and http://nodal.sourceforge.net/ (04)
Now that I have looked at the site, this does ring a bell. (05)
% > I have been pushing around the idea of a distributed/replicated
% > global file system (I am still searching for one)
% > and coming up with a tuple space file system, which would be nodal. (06)
% > What is the underlaying structure that you need? (07)
% The later versions of the ReiserFS provide a lot of what's needed.
% I talked to Hans Reiser last year and from what I understand, despite
% them not receiving much if any more money from SuSE linux, I think
% they're going ahead with the plans for enhanced node-level metadata.
% I haven't followed this in about a year, so I'm not sure. But some
% combination of ReiserFS at the file system level, coupled with ideas
% from Lee's system would probably do it. NODAL is operating at a
% higher level than ReiserFS, and I think you'd probably want to be
% down at the FS level to do any of this efficiently. It's a radical
% idea with lots of potential. (08)
% I've been using ReiserFS as my main file system on linux for
% several years, and I'm not a FS hacker, but that's probably where
% I'd be hacking if I was. (09)
Ok. Clearly I need to learn more about reiser. (010)
John Sechrest . Helping people use
. computers and the Internet
. more effectively
.
. Internet: sechrest@peak.org
.
. http://www.peak.org/~sechrest (011)
--
This message is archived at: (012)
http://collab.blueoxen.net/forums/cgi-bin/mesg.cgi?a=tools-yak&i=200402070040.i170ebY17191@jas.peak.org (013)
|