In some environments where disk replication is in use (such as with Veritas Replication Exec , Veritas Volume Replicator option for Storage Foundation , Microsoft DFS and so on), you need to take extra consideration to prevent excessive replication traffic.
When data replication is in place, frequently it is used to replicate data across a WAN for disaster recovery purposes. As WANs are relatively slow (compared to LANs), it's beneficial to avoid or reduce replication traffic where possible.
When data which is being backed up is replicated, if you use a backup method which resets the archive bit (such as Full or Incremental), then a write occurs to each file backed up. This is relatively small, so will create replication traffic, but not excessive amounts. If you use differential backups, then these writes should be avoided. Obviously this should only be an influencing factor in your backup strategy decision.
If you use the Advanced Open File Option (AOFO), you should try to avoid the AOFO cache file being placed on a replicated volume. By default the cache file is placed on a drive different from that being backed up. Suppose you have a server with two drives, C: and D:, and D: is used for data and replicated. If AOFO is used for the backup of the C: drive, then the cache file must be placed on D:. This may create a large amount of replication traffic. Ways around this might be to avoid using AOFO where possible, or to add another drive to the server, and using the Advanced Open File Option Wizard (Tools, Wizards, Advanced Open File Option Wizard), set this drive to be used for the AOFO cache.
Replication is sometimes used to replicate Backup-to-disk folders. If the replication capacity is available, this can eliminate the need for backup tape vaulting. Because large files are used for the B2D files, it is best to use replication software which works on a block-by-block basis, rather than at a file level. If you have to use file-level replication, it is best to keep the B2D file size relatively small, otherwise if you set the file size to 100GB, then backup to an existing file, you'll use 100GB of replication traffic.