Posted by Gareth Cutter on 8th June 2009 to
Backup Planning It can be difficult to imagine what a data loss scenario will be like; it might be all of your data is lost, or only part of your data is corrupted and you have to find those missing parts urgently. If you’re using a disk or tape-based backup system and the files number in their thousands, sorting through these records can be a time consuming and frustrating process. Does this really qualify as data backup, or is it just a cache of disorganised copies?
According to several prominent bloggers, data backup can only be called such when it meets the following strict criteria:
- The backup resides on a different array and / or location from the source data
- Creation, agency and dispensation of these copies is managed by an application
- The backup moves off-site at some point during the backup procedure
- The data is able to recover either in whole or in part
- Recoveries seek a coherent point in-time representation
- The existence of the data backup does not effect the primary set
If your current data backup procedure doesn’t satisfy all of the above criteria, it probably isn’t adequate for the job. It’s your responsibility to ensure business, employee and client data is safe, or face the potentially disastrous consequences of data loss.
