[Gluster-devel] Snapshot backend (was Re: EAGAIN/EBUSY handling in glusterfs)

[Niels de Vos]

On Wed, Jan 23, 2013 at 10:33:49AM -0800, Joe Julian wrote:
> 
> On 01/23/2013 01:34 AM, Shishir Gowda wrote:
> >One of the possible scenarios is someone taking a lvm snap of the backend.
> As an aside, lvm snapshots are good for one-off temporary saves, but
> are completely useless as an interval snapshot. Each one takes up a
> ton of ram and increase boot-up times very significantly. I would
> recommend focusing on btrfs for any snapshotting backend.

The new LVM/device-mapper thin-provisioning should remedy quite these 
problems. I have not yet tried this myself, but it looks promising.

- http://fedoraproject.org/wiki/Features/ThinProvisioning
- http://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git;a=blob;f=Documentation/device-mapper/thin-provisioning.txt