We to are users of EMC disk and have purchased FDR SOS.

Though we have had SOS for a while I never implemented it nor do I intend to do
so because a couple of limiting factors.

I can only see the SOS backup option for DR purposes since a single file restore
is not possible unless you are willing to restore the whole disk backup to
another location (maybe a BCV) and then restore from that.

Assuming I am using SOS as a DR solution ....

In the case of a server being rebuilt onsite and the EMC disk is still in one
piece this option is perfectly valid.  However, I require that the same backup
method / media  be suitable for this scenario as well as total recovery in the
event of  catastrophe, i.e. someone parked a jumbo on the computer centre.

The problem with the SOS solution for a catastrophic DR scenario  is that my
alternate site must have another EMC disk subsystem configured exactly the same
as the original or at least suitable for DR requirements.  The problem here is
not the break up of the disk config but the cache, and host adapters and the
actual disk sizes (does the DR disk have 47Gb  or 18 Gb drives).

I can not see how a DR service provider can cost effectively cater for vast
number of possible config variations.

So what is next?  You could purchase another identical disk and have it sitting
ready at the DR recovery site. But if you are going to do this then why not have
all the data or at least critical data on your primary disk remotely copied
utilising EMC's SDRF and Timefinder options.

This seems a very good option however it the negates the need for SOS since the
DR situation is covered by a remote copy. The data can be further protected by
using timefinder at the remote and / or local site the have a fast restore
option  to the latest version which would make up over 90% of your restore
requests.

The Timefinder copy potentially be used as the ADSM backup target rather than
the real data therefore negating the need for backup downtime.

If the hype of IBM's ESS (shark) comes in fruition then the cost of this level
of disk is subject more downward pressure making the remote disk solution more
realistic.

My guess is that  ESS  will even offer more features for Open systems backups.
Since the ESS already has a couple of RISC processors in it I cannot see why
they will not put in a third to act as an ADSM server. Once virtual is up and
running and therefore Flashcopy then if my guess (or hint to IBM as it may be)
is correct then  ADSM could not only offer instant backup within the same disk
array but it will involve no data movement until it is required to go to tape.




Peter Griffin
Zurich Financial Services Australia
peter.griffin AT zurich.com DOT au







--------------------
"This email is intended for the named recipient only.  The information contained
"This email is intended for the named recipient only.  The information contained
in this message may be confidential, or commercially sensitive.  If you are not
the intended recipient you must not reproduce or distribute any part of the
email, disclose its contents to any other party, or take any action in reliance
on it. If you have received this email in error, please contact the sender
immediately.  Please delete this message from your computer."
--------------------
=========================================================================