I have been doing some reading about database backup and restore.

While I have some understanding about full / differential database backup and restore + subsequent log backup and restore, I can not comprehend the usefulness of DATA /FILEGROUPS backup and restore.

I am not asking about the syntax, but rather its usefulness and situations under which regular full database backup + differential database backup + log backup + tail-log backup would not be sufficient or inappropriate.



Well, here is a simple example. You have a large database, but most of the data is history. You go ahead and partition out the historical data onto read only filegroups. This information is used on a regular basis for reporting but not for the actual application. It is, lets say, 90% of your database.

If there is a problem you can restore the read/write portion much more quickly to get the application up and running. Then once your application is functioning you can restore the read/only portion (which of course will take much longer).

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.