Bill and Paul thanks for the input.

The full requirement is for a full monthly backup kept 13 months and a yearly snapshot 
that is to be kept indefinitely.

I did an analysis of 4 options using my current production servers and data volumes

Monthly archive,
Monthly incremental to "monthly" node
Tape backupset
Backupset to sequential disk files of 1 GB size  which are then migrated using HSM

One key problem with backupsets is the time that they take, and you're not supposed to 
perform any backups or expiry whilst they are being created.   Given the tape usage 
profile on my system that means new tape drives, and costs slots in the 3494. The 
backupset with migration option was surprisingly even more expensive - again bigger 
drive requirements in order to migrate some of the backupset data files whilst the 
backupset is still being created .

The problem with the monthly incremental is that the yearly snapshot has to be taken 
by a different method. Although this option did cost out cheaper, it wasn't by much.

The Monthly archive costed out to be second least expensive, and the same method can 
be used for monthly and yearly snapshots. Its also driven from the server side and so 
requires less fiddling on each client. Using the KISS principle, this is the winner.  
The big downside is, of course, the database size increase.  I can cope with the 
archive network  load provided I split the work over the 8 days of 4 weekends.

So, any input on configuration management?

Steve.

>>> [EMAIL PROTECTED] 29/06/2002 0:08:35 >>>
Why not use backupsets? I know, you need 1 tape per server for each
backupset as opposed to putting all the new monthly backup data in its own
pool and filling the tapes.

We thought about a couple other ways around this...virtual volumes to
another server and having a FILE device class for the backupsets and then
archiving the backupset files to TSM tape.

For the virtual volumes, you need another TSM server license and full
support for backupsets as virtual volumes is not there. The RECONCILE
VOLUMES command does not recognize a backupset as a valid volume type and
deletes the virtual volume, but not the related backupset entries. It works,
but you have to remember to NOT run the command.

For the FILE devclass, you need enough disk space to hold a backupset. Plus
to restore from the backupset, you would need to retrieve the archive that
contains that backupsets' files first.

Just some $.02 worth...

Bill Boyer
DSS, Inc.

>>> [EMAIL PROTECTED] 29/06/2002 11:59:36 >>>
You said backup, did you mean a full incremental once a month?
If so, the easiest way to do that is create a second node and second stanza
in your dsm.sys for the server with a different policy domain/management
class and use the -se option with a private dsm.opt for your include/exclude
list.  Mark the copygroup with mode=absolute.  Use a separate storage pool
if you would like to send just those offsite.

Paul D. Seay, Jr.
Technical Specialist
Naptheon, INC
757-688-8180






**********************************************************************
This e-mail, including any attachments sent with it, is confidential 
and for the sole use of the intended recipient(s). This confidentiality 
is not waived or lost if you receive it and you are not the intended 
recipient(s), or if it is transmitted/ received in error.  

Any unauthorised use, alteration, disclosure, distribution or review 
of this e-mail is prohibited.  It may be subject to a statutory duty of 
confidentiality if it relates to health service matters.

If you are not the intended recipient(s), or if you have received this 
e-mail in error, you are asked to immediately notify the sender by 
telephone or by return e-mail.  You should also delete this e-mail 
message and destroy any hard copies produced.
**********************************************************************

Reply via email to