Difference between File recovery points and Application Recovery Points RRS feed

  • Question

  • Hi,

    I  have DPM 2012 SP1,

    When protect system state and bare metal we got option of File recovery points and Application recovery points

    What is difference between them. can some one explain with Practical example.

    Can I set schedule  to both run at same time.

    Thanks in advance.

    Sunday, September 29, 2013 2:09 PM

All replies

  • See for a explanation of the differences, but essentially File recovery points handle data which can simply be backed up, eg flat files, documents, images etc, things that don't required interaction with the relevant application in order to backup locked / in use data.

    Application recovery points handle data that requires alternate means to backup the data. For instance VSS for the system state, talking to SQL to backup SQL databases, talking to Exchange to backup message stores etc. The exact method used in each case depends on what is being backed up.

    Since this option is setup at the protection group, and that can contain data in both categories, you get both options listed on that screen.

    Sunday, September 29, 2013 2:43 PM
  • Hi Keith,

    Thanks for the explanation.

    Can this be both scheduled to run at the same time? as per the original question above.

    What would be Advantages or disadvantages of doing so ? Or should it be run at different times?

    Please advice.


    Tuesday, December 17, 2013 10:23 PM
  • Yes, those details are configured within the protection group, and you can have a combination of both within a single protection group. For instance, if you had a SQL server being backed up you could have a protection group for that server, and within it you might have C:\, E:\, System State, SQL\database1, SQL\database2 etc listed.

    DPM works out whether the resource being backed up needs a file or application recovery point automatically.

    For a small setup I'd likely go with having all of them in one, but with larger setups you can run into problems. In practice we tend to have a protection group for the drives & system state for a server (or a small collection of servers), and then a separate protection group for all the SQL databases or exchange mail stores etc. This then lets you tailor the schedule accordingly, for instance the mailbox store's likely to change regularly so you want more frequent backups, whereas the file system won't change as much.

    There is a technical reason for this being better, but unfortunately I can't remember what it is now. We started putting everything together, ran into problems, split things up and then found that was the recommended way, so have done it that way since, especially since it's run smoothly ever since.

    Tuesday, December 17, 2013 10:48 PM