It’s no wonder that people in our industry have such a hard time figuring out what platforms can actually perform what features. Probably one of the most confusing is the phrase “Point In Time Recovery”. Both databases and storage (and virtualization) people use this phrase and it means two VERY different things depending on who is saying it.
What database people use this phrase they mean that they want to be able to restore the database to any point in time.
Storage and Virtualization People
What storage and virtualization people mean is that they can restore the system to the point in time that the snapshot was taken.
The test that I always provide people with is that, if I can a phone call today saying that the database needs to be restored to January 17, 2015 at 1:34:14pm can your solution meet that requirement without knowing about the requirement back on January 17th. If the answer is yes, then you have point in time recovery, if the answer is no then you do not have point it time recovery.
The problem is really just a phrasing issue more than anything else. In reality both the database backup solutions and the storage/virtualization backup people (think array snapshots, Veeam, etc.) can do point in time restores. Database backups done with the transaction logs can restore to any point in time, while the storage and virtualization folks can restore to pre-defined points in time. While the phrases are the same, the actual meaning behind them is VERY different.
When talking to storage and virtualization folks who want to take away the ability to do backups from the database team we need to get a clear understanding between everyone involved as to what is expected on the database backups, and from a technical perspective not from a marketing / sales / English language perspective.
Contact the Author | Contact DCAC