Unstructured Data Backups in Scale-Out Repositories

If you use a scale-out backup repository as a backup repository for storing unstructured data backups, Veeam Backup & Replication distributes every 1 GB of backup data across all suitable performance extents in the round-robin manner. This allows balancing both storage capacity and load, including the load on gateway servers for object storage extents.

An extent is considered suitable if it meets the following conditions:

  • The extent is available.
  • The extent is not in the Sealed mode.
  • The extent has enough free space to write 1 GB of data and the metadata copy.

Tip

To avoid writing unstructured data backups to specific extents, put these extents into the Sealed mode. Veeam Backup & Replication will not write new data to sealed extents.

The backup data is accompanied by the metadata files. To provide an opportunity to restore data written to extents remaining available in case one or two of other extents are unavailable, Veeam Backup & Replication triplicates metadata when writing it to extents. Thus, every extent keeps metadata for data stored on this extent (in folder meta) and replica of metadata for data stored on adjacent extents (in folder metabackup). Even if two of three adjacent extents are lost, you can restore backup data stored on the third extent.

Note

Consider the following when storing unstructured data backups on a scale-out backup repository:

  • The metadata redundancy approach protects against two extents out of three being completely lost.
  • The metadata redundancy approach allows protecting metadata, not data.
  • An unstructured data backup job consumes 1 task slot on every extent of the SOBR per each file share added to the job.
  • The backup file placement policy set for the scale-out backup repository, as described in section Backup File Placement for Performance Tier, is ignored.

Unstructured Data Backups in Scale-Out Repositories

Page updated 2026-11-12

Page content applies to build 13.1.2.57