Taking backups and never having to use them, is often considered as a waste of time. But whenever you get a situation where a backup has to be restored partially or completely, you will appreciate its value.
Have a Strategy
That is why it is always recommended to have a good recovery strategy. It should not only be defined once, but also regularly reviewed and adapted to the current needs and to the experiences made when recovery has been required.
This recovery strategy should cover both backups and the possible restore scenarios. It should comprise recovery on the same system, but also on another system. Even a disaster backup and recovery should be considered, where the whole environment can be recreated from scratch on another system, for example in an IBM Backup and Recovery Center.
Test Backup And Restore
The procedures and the backups should also be tested, for example a recovery should be tried from the backups actually taken either on the same system, or if possible on a different system. If these tests are performed, they might reveal a need for changes in your archiving and recovery procedures, so that you will be able to benefit from these in case you need them.
Keep Several Backups
Keeping different sets of backups, and if applicable, log archives, may give you the possibility to return to an earlier point in time than the latest archive. This can be helpful, for example:
Starting from that older archive, a table can be restored individually with forward recovery and this can be stopped at exactly the point in time that you decide, where the data reflects the correct information that you need.
In addition to regular backups (see page ***), consider to include backups in your strategy: