About backing up data to Amazon Glacier vault
When a NetBackup backup job is run to back up data to use the storage class, the data is stored in the vault as a set of archives. The metadata is stored in an S3 bucket as storage class objects as well as in the Amazon Glacier vault as set of archives.
Considerations:
If a backup fails due to network issues, the partially backed up data may reside in the vault and hence occupy storage space.
It is recommended using the LIFECYCLE storage class for moving data from other cloud storage classes to Glacier. See Protecting data using Amazon's cloud tiering . However, to move data from other cloud storage classes to GLACIER_VAULT, you must host a cloud media server and duplicate data through it. This step is done to avoid data-out cost. This workaround also applies to moving data from GLACIER_VAULT storage class to other cloud storage classes.