Table of ContentsLibraryView in Frames

How the NearStore secondary system processes the data

It is important to understand how the NearStore secondary storage system processes the data.

Backup data transfer: Each backup data transfer to the NearStore secondary system consists of two streams: a standard NetBackup tar stream that contains header and file data, and a metadata stream that describes the contents of the tar stream.

The tar data is written to the NearStore secondary storage system and stored on disk. Keeping the tar stream on disk allows restores to be initiated by NetBackup from the NetBackup Administration console.

Using the metadata stream, the files and directories inside the tar stream are unpacked into the WAFL file system in Data ONTAP. The unpacking process creates files and directories in Data ONTAP as the tar stream is written to the NearStore secondary storage system. The file data in the unpacked file system does not consume additional blocks because it shares all of the tar stream data blocks.

Context-based deduplication: As Data ONTAP unpacks file data, it performs context-based deduplication. Context-based deduplication eliminates common blocks between the current tar stream and all previous tar streams; therefore, only new or changed data blocks in the new tar stream are written to disk on the NearStore secondary system.

Snapshot copy creation: After all of the data is written to disk and all the data inside the tar stream is unpacked into an active file system, the backup is captured to a Snapshot copy. In this way, backups of the primary file system are captured to unique Snapshot copies that you can browse if you need to restore a previous version of files or directories.

Data ONTAP optimizes the number of Snapshot copies consumed by coalescing backups for different polices into a single Snapshot copy. For example, if three backups for different policies finish before the next update to any of the backups, Data ONTAP coalesces the three backups into one Snapshot copy.

Restoring data: Before Data ONTAP 7.2, you could only restore data from SnapVault for NetBackup volumes by accessing the tar stream using the NetBackup administration console.

Data ONTAP 7.2 and later enable you to access the unpacked file system using either the NFS or the CIFS protocol after a backup finishes. This allows a second method to restore data, using “drag and drop” functionality to copy files and directories to the source from the backup.