Hello Adam,
An idea might be to define for each client two node definitions,
NODENAME and NODENAME_ACTIVE. NODENAME would be your normal daily
backups keeping x versions for y months. NODENAME_ACTIVE would be a
separate incremental backup of the same client that you backup to a
management class that keeps one version of the backup (no inactive
versions). This has the disadvantage that you are doing two incremental
backups per day of the same client, the advantage is that the
NODENAME_ACTIVE restore would only search through active data, therefore
search times and number of mounts would be greatly reduced (search times
should be next to nothing for a full restore.)
I haven't implemented this, so I don't know how it works in practise,
but it might be an idea for you. If you are currently retaining backup
versions for a month or more, then it should use less library space than
you use now with full backups each week.
Regards,
Paul.
-----Original Message-----
From: ADSM: Dist Stor Manager [mailto:ADSM-L AT VM.MARIST DOT EDU] On Behalf Of
Adam Sulis
Sent: Thursday 13 January 2005 16:56
To: ADSM-L AT VM.MARIST DOT EDU
Subject: TSM DR Exercise Restore Ideas
Hi All:
I am looking for suggestions to improve efficiency in our Backup
strategy:
Current:
We backup 200 nodes to a DASD diskpool (large enough for one day's
backup),
then migrate to a VTS, with an offiste copy going to 3490 (uncollocated,
of
course). To improve day-to-day restores, we implemented a weekly
full-backup
for our nodes. This means that restores come from a maximum of the last
1
week's worth of tapes (only need the active versions of the files).
Also,
these full backups make our Disaster Recovery restores easier to work
with
(when we get to the DR location, we call for tapes from the past
weekend,
and all incrementals since. The problem is that we only have the
city-to-city bandwidth to offer weekly full backups to the nodes
participating in the DR plan...
Proposed:
What I would like to do is remove the VTS, have the backups direct to
DASD,
then migrate to a collocated 3490, with an offiste copy same as before.
Having done this, we would have improved backup performace for all
nodes,
and provided great restore performance for all nodes.
Problem: Without weekly full backups, the offsite tapes required for a
restore would grow to a huge number (Operating System files would be on
a
very early tape, and every tape generated since would have incremental
data
for that node). Is there any way to make a shorter list of tapes
required
for DR? How is everyone else dealing with the offsite tapes? Again, we
are
only interested in the "active versions". I could aggressively reclaim
the
offsite pool to create newer tapes, but that's a lot of TSM thrashing
for
little gain (still cannot say to the vault "send me all tapes since last
Saturday". Daily backupsets do not work within the DR plan, as far as I
can
tell. (Unless I can generate a backupset to remain inside TSM, which
would
get copied to the offsite daily, and still be available at the DR
site...
I must be missing a command or a concept - how is everyone else doing
this?
Any thoughts appreciated.
Adam Sulis
DNIS6-5
2nd Floor, Tunney's Pasture L002
Tel (613) 998-9093
Sulis.AJ AT forces.gc DOT ca
Any e-mail message from the European Central Bank (ECB) is sent in good faith
but shall neither be binding nor construed as constituting a commitment by the
ECB except where provided for in a written agreement.
This e-mail is intended only for the use of the recipient(s) named above. Any
unauthorised disclosure, use or dissemination, either in whole or in part, is
prohibited.
If you have received this e-mail in error, please notify the sender immediately
via e-mail and delete this e-mail from your system.
|
