Data Pruning from Archive Cloud Storage

Data is pruned from Archive Cloud storage as follows:

  • For non-deduplicated data, the data copy is aged using the job-based aging rules. After a job completes, the copy has a retention time, and when there are no other holds on the data, the job-contents will be deleted from the Archive Cloud storage.
  • For deduplicated data, the DDB must be sealed periodically (For example, every six months or every year) to make sure that the data is pruned, as micro pruning (deleting individual jobs) is not supported on Archive Cloud storage.

    Micro pruning is an extremely slow and impractical process on Archive Cloud storage products. This is because the pruning process will modify the individual objects that are referred to by multiple backup jobs (for deduplication). To modify such objects, the objects have to be brought back to the non-archived cloud storage first, modified, and then put back into Archive Cloud storage.

    Hence, it is recommended to seal the DDB to create smaller DDBs that can be macro-pruned. Macro-pruning means the deletion of a job without the DDB. Essentially all the data no longer needs to be modified, as the DDB is no longer active (sealed). During the macro-pruning process, when the DDB is sealed, the DDB will be aged when all the backup jobs associated with that DDB meets the retention rules. Once the DDB is aged, a pruning request is sent to delete the jobs, and the corresponding jobs will be cleaned up from the Archive Cloud storage.

Last modified: 12/6/2019 8:21:17 PM