Regarding Deduplication

[adamwoodinfo]

Whether server side deduplication in BACKUP EXEC is a single user based or is based on all the data that is available in a media server from different users?

Any idea on how deduplication is done? basic functionality?

Thanks,

Adam

[dedupe-works]

The account used for backups should have access to all data for backup purposes.

This is independent of deduplication.

Deduplication in Backup Exec uses a fixed segment length of 128KB to create fingerprints of the data stream received for each file. Any fingerprints that match the fingerprint cache are considered redundent and are not stored as new data on the hard disk. At that point a new link to the existing fingerprint is created.

At this time, there is no scheduled deduplicating of data. This is done on the fly while the backup is running.

[adamwoodinfo]

Ya. That account uses all the data for backup process. But what my doubt was, suppose there are 100 users provide deduplication backup to a same media server then, when an user try to give a deduplication backup to the server means, whether deduplication process is done based on the data that was previously backed up from the same user or it consider all the data within the media server from all the 100 users for deduplication

Adam

[dedupe-works]

The deduplication process is performed against data store on the media server.

The data previously backed up is compared to other data in the data store, regardless of the user.