Remote Seeding with Catalogic DPX OSS

For remote data protection (for example, of a branch office) it can be a challenge to perform an initial base backup if the source primary system contains a large amount of data and is to be backed up over a slow communication link.

The Catalogic DPX Block Data Protection remote seeding function solves this problem. It allows for remote block-level backups over very slow connections that would normally take a prohibitive amount of time. The process involves a full backup to a locally attached tape or portable hard drive. The media is then transferred via physical delivery (FedEx, UPS, etc.) to a site running the Catalogic DPX Open Storage Server. This full backup is then restored to the Catalogic DPX Open Storage Server, after which incremental backups continue over the slow internet connection. Once the first incremental backup is complete, the data will be available for recovery.

Note the following requirements.

• The Catalogic DPX remote seeding procedures described here apply to Block backup to the DPX Open Storage Server.

• The client node (also referred to as the source or primary node) must be Windows.

Consider the following example. From the company headquarters in New York, an administrator wants to back up a node in a small California office to a server at the headquarters. Over a robust Internet connection, such a backup is handled using normal backup procedures, regardless of physical location. However, the California office has a link with a very slow upload speed. A conventional backup ties up this connection for several days and probably fails due to the high likelihood of service interruptions over such a long period. Instead, the administrator uses the Catalogic DPX Remote Seeding process.

Using the management console from his office in New York, the administrator initiates a seeding backup job, chooses the remote node that they want to back up and a remote backup device (at the same location as the source node, in this case California). Once the full backup is complete, the administrator has the media shipped to New York, where they initiate a seeding restore job and restores the media to the Catalogic DPX Open Storage Server. Once that restore job is complete, Catalogic DPX continues running incremental backups from the source node in California to the Catalogic DPX Open Storage Server in New York over the slow connection.

The backup media is typically tape, but a portable or removable hard drive – even a laptop – can be used if storage capacity is sufficient.

The following diagram illustrates a sample scenario:

Stage 1.

A full (base) backup to directly attached media followed by transporting of media to the Catalogic DPX Open Storage Server location.

Stage 2.

Restore of base backup to the Catalogic DPX Open Storage Server.

After Seeding.

Automatic incremental block-level backups over slow internet connection.

Note that Stage 1 (backup) and Stage 2 (restore) are initiated from the management console task panel, where Define Seeding Job displays if the licensed Catalogic DPX Open Storage Server is present in the Enterprise.