The volume of data created and stored by organisations continues to grow at amazing rates. Balancing the need for application availability and performance with that to capture and store back-up copies of the data, many organisations that for years focused on tape as their primary back-up destination are now turning to disk as the initial (and sometimes final) destination, a process that has become known as disk-to-disk (D2D) back-up.
There is a whole host of reasons behind this shift such as the dramatic decline in the cost of hard disks, the steep rise in their capacity, and the much better read/write performance of disk arrays compared to tape libraries. This extra performance is particularly valuable if an application must be paused or taken offline to be backed-up, as the amount of downtime can be minimised.
Disk can be incorporated into the back-up process in many ways, from virtual tape libraries through snapshots to continuous data protection, and each method may suit some applications or user requirements better than others. An organisation may even use more than one D2D back-up scheme in parallel in order to address different recovery needs.
There is more to using disks for back-up than merely speed. A big advantage of disk over tape is that disk storage is random-access, whereas tape can only be read sequentially. The use of disk technology has also enabled another key innovation in back-up: de-duplication. This is a compression or data reduction technique that takes a whole set or stream of data, looks for repeated elements, and then stores or sends only the unique data.
Replication and mirroring
Disk is also used in many other forms of data protection such as data replication and mirroring, although it is important to understand that these are not the same as back-up. They protect against hardware failure or disasters, but they cannot protect against data loss or corruption, as they offer no rollback capability.
However, the combination of replication with other disk-to-disk back-up solutions can provide effective answers for protecting data generated and stored at remote offices where restricted bandwidth makes a more traditional back-up strategy such as daily incrementals inappropriate.
In order to select the correct solution for their environments, companies must first understand the recovery time and recovery point objectives that enable them to meet service level agreements and maintain the availability of key business systems. It is these two important factors that should define the requirements for disk-based data protection, and so help identify which solutions will provide the organisation with the level of protection they need.
Andrew Brewerton is technical director at BakBone Software
This was first published in July 2009