Re: Data De-duplication

[Ray Van Dolson <rayvd@bludgeon.org>]

I lost the original post so I'm jumping in at the wrong thread-point :)
Someone mentioned that the primary usage of de-dup is in the backup
realm.  True perhaps currently, but de-dup IMO is *the* killer app in
the world of virtualization and is a huge reason why we're picking
NetApp at work to back our NFS VMware DataStores.  We easily see 50%
savings in space.

I know of only one other production filesystem implementation of
data-dedup -- GreenBytes has it in their ZFS-based storage product.

I'm not sure why this hasn't caught on, but as soon as a solid and fast
implementation of it exists in the Linux world I really think it can
catch on for VM datastores.... I know we've hollered at Sun as to why
they haven't rolled it out for ZFS yet!

Anyways, I know it's on the roadmap, just like throwing my $0.02 once
in a while on how big a feature I think this could be......

Great job all :)
Ray