Node Replication and Disaster Recovery

[snowywar]

Hi Guys/Girls,

we seem to be moving TSM more into the world of dedup technology and node replication and I just wondered if anybody has come across any good docs, redbooks or tech notes that looks at the options and more importantly the pit-falls.

My concern is that we can copy all node data from one TSM instance to another but what can we do to backup the TSM DB.

After a disaster I don't want to rebuild the TSM instance from scratch and then configure the solution from scratch.

In a world without tape.

Your thoughts are welcome.

regards

Richard

 

[DanGiles]

You have 2 main options (and various sub-options which we won't get into right now). The first is to use remote vaulting of the TSM servers. The destination at the remote end can be any form of media - tape, disk, deduped disk... In case of disaster, you will have to rebuild the TSM server, but not from scratch. The other is to use Global HACMP; much more expensive, but in case of disaster you can be up and running in minutes.