Recovery point objective
A recovery point objective (RPO) describes the data loss tolerance for a business function or application. For example, the payroll department may have an RPO of 24 hours for the PeopleSoft payroll module, meaning that in a worst-case situation, they can afford to lose and/or be forced to recreate 24 hours of data.
Let’s explore a data loss scenario in practical terms. Company X uses a tape backup solution for PeopleSoft. Once a week, a full backup is performed, and each night (at approximately 2:00 a.m.), daily incremental backups are performed.
At 8:00 a.m. each day, the tapes are moved off-site, where they reside for one month (at a minimum).
Let’s examine three examples in order to demonstrate how to calculate data loss potential based on the process used by Company X (assuming the last backup was run on November 3 at 2:00 a.m. local time, and the tapes were taken off-site the following morning at 8:00 a.m.).
• 10:00 a.m. on November 3 – data loss potential of eight hours (10:00 a.m. minus 2:00 a.m.)
• 3:00 p.m. on November 3 – data loss potential of 13 hours (3:00 p.m. minus 2:00 a.m.)
• 6:00 a.m. on November 4 – data loss potential of 28 hours (even if a tape backup was performed at 2:00 a.m. on November 4, the tapes are still on-site; therefore, the backup media is unavailable if the data center was destroyed or unavailable. As a result, the November 3 media would have to be used.)
Assuming an RPO of 24 hours, management may or may not be willing to accept the risk of this current backup process.
Recovery point objective gives an indication of efficiency