Given the data volumes being handled and the likely complexity of any viable backup strategy, it is important to automate as much of the backup process as possible. For example, by the use of stacker or silo technology, large backups can be accomplished without operator intervention. This minimizes the chance of error, and avoids any potential delays. It is equally important to automate the scripts that run the regular backups. These scripts need to be sophisticated enough to report any problems and to either deal with them directly or allow the job to be restarted once the problem has been fixed.