replicated file system?

David Miller david3d at gmail.com
Tue Feb 28 15:16:04 EST 2012


On Tue, Feb 28, 2012 at 2:50 PM, Ken D'Ambrosio <ken at jots.org> wrote:

> There *are* distributed filesystems -- btrfs has ceph, which has come a
> long,
> long way. Lustre and Gluster also come to mind.  Caveat: I've not used
> these,
> but I know folks who have, and I believe they'd fit your bill.
>
>  -Ken
>

Of those I'd recommend Gluster as it's relatively easy to setup and use.
 It's also a userspace filesystem so you can avoid having to deal with
custom kernel modules that you have to recompile every time you update your
kernel.  I think I'd look at Gluster over DBRD.  But options are good to
have.

I'd avoid Lustre is a nightmare to setup and if I recall correctly it
really needs more than 2 nodes it needs to have dedicated metadata
server(s).

Ceph looks interesting but of these solutions it's the newest and currently
relies on BTRFS which is getting there but I don't know that I'd be willing
to use it in production just yet.
--
David
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.gnhlug.org/mailman/private/gnhlug-discuss/attachments/20120228/e91d2de3/attachment.html 


More information about the gnhlug-discuss mailing list