Coda File System

Re: [kind of offtopic] Re: CODA Scalability

From: Ivan Popov <pin_at_math.chalmers.se>
Date: Thu, 22 Aug 2002 17:42:33 +0200 (MET DST)
On Fri, 23 Aug 2002, Nick Andrew wrote:

> a cell and the new server's available disk is equivalent
> to the existing server's disk, from the point of view of

Right.

> clients. One must still move volumes from server to server to
> balance utilisation, but at least this can be done without
> downtime.

It is unfortunately not trivial to move molumes in Coda yet.

> The Network Block Device (NBD) driver allows a remote host's block

I tested once the following setup:

nbd on multiple machines, serving the disk space to a "filesystem server",
that was going to run raid5 over the network "disks" and then put an
encrypted filesystem on it, thus protecting the data integrity and privacy
from corrupted disk space servers. Then you can export that filesystem in
any way, with NFS, Coda, InterMezzo and so on.

The project showed that raid is not suitable for "disks" that
 - can silently contain wrong data
 - can disappear and then come online again, without
   manual intervention to "replace" them
 - can become arbitrarily slow (depending on load and network)

A rewrite of both nbd and raid/lvm would be necessary, so I abandoned the
project. Otherwise it is pretty workable.

Regards,
--
Ivan
Received on 2002-08-22 11:43:39