[dm-devel] [RFC] DM Snapshot scalability - Common snap store approach

[Mike Snitzer]

On 10/9/06, Vijai Babu Madhavan <MVijai at novell.com> wrote:

> Quick summary
> =========
>
> Various approaches that are being considered/prototyped to solve the
> scalability problem with
> DM snapshots are described here. Currently, I like the 2.d approach
> (using a single cow device
> for all snapshots of an origin, with a combination of exception
> stores). If you don't want to
> read about other approaches, jump there directly. I would like to know
> about other ways
> to solve this problem, and comments about these approaches.

A shared exception store was included as part of the cluster snapshot
prototype that Daniel Phillips worked on some time ago:
http://sources.redhat.com/cluster/csnap/

See section "1.3  Multiple Snapshots" and "5 Server Implementation
details" in csnap.ps.  And csnap.c in the csnap.0.0.1.tgz

This code is quite cluster oriented but it does provide a good amount
of detail on how one might implement a shared exception store.  I
kernel-based solution could come later...

It would be _very_ nice to improve the current dm-snapshot with shared
exceptions; I for one am eager to see continued
discussion/implementation of this.

Mike