"ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU> wrote on 09/22/2009
10:16:54 AM:
> Hi!
> Before they were taken over by EMC, some guys from DataDomain
>were visiting us a few months ago. They presented their DD boxes,
>which offer hardware based dedup, compression and defragmentation.
> They told us about customers who allocate a DISK (not FILE!)
> storage pool in a DD box and just leave the data there in one huge
> diskpool. Because the box does his own backend fragmentation, the
> front-end fragmentation is just virtual and thus does not impact
>TSM performance.

> It does sound too good to be true.

Yea . . that does sound too good to be true.

For a DISK pool on DD, they must be talking about using the NFS interface.
In this case, the DD would be just NFS mounted filesystem to TSM.
So the file volumes in the DISK pool would be files to DD.  DD
could defrag the file volumes, but they cannot defrag the contents of
the file - TSM fragmentation.  The file volumes would themselves be
optimised in
the DD, but not the contents of the file.

FILE devices on a DD NFS mount would be files to DD, and would be
optimised when the DD did a defrag.  But there would still be 2
levels of defrag:  DD and TSM (reclamation).

But then, you are talking about compressed and deduped data anyway.
I wonder if any kind of sequential access is really sequential at all!

(my head is hurting)

-----------------------------------------
The information contained in this message is intended only for the
personal and confidential use of the recipient(s) named above. If
the reader of this message is not the intended recipient or an
agent responsible for delivering it to the intended recipient, you
are hereby notified that you have received this document in error
and that any review, dissemination, distribution, or copying of
this message is strictly prohibited. If you have received this
communication in error, please notify us immediately, and delete
the original message.