Access and Feeds

Storage: Where is Your Data?

By Dick Weisinger

Continuous data protection (CDP) is a method for tracking changes to your data by versioning data every time a save is done.  In that way, unlike traditional backup, it is possible to restore data to any point in time.  Traditional backups are limited to backup snapshots from specific staggered points in time, typically daily.  CDP happens continuously so there are no scheduled backup times.

CDP data changes are captured and stored to a separate storage area, usually performed by asynchronous writes to another machine on the network.  CDP enables the possibility for viewing the chronology of data changes.

CDP is different from mirroring because a mirror image of the data will be current with the current version of the data.  If something is deleted or corrupted in the current data set, a mirrored image will be of no use — mirroring is useful in the situation when the main data source becomes unavailable and the mirror is immediately ready to be brought on-line.  CDP allows you to select any point in time from which you can reconstruct the state of your data.

Gartner estimates that the entire backup and recovery market to be about $3 billion globally.  They estimate that about 18% of organizations use CDP on their backend servers to safeguard critical application data, and about 4% of companies are using CDP to backup data from client machines.  More organizations seem to be switching from traditional backups to CDP, but many companies are finding that traditional backups are good enough for many scenarios while CDP is more useful in situations involving critical data.

CDP can also become a drag on performance and resources.  “You’d be ill-advised to apply CDP to all types of backup because the level of system overhead is quite high,” he said. “This can lead to bottlenecks, and the underlying storage requirement can also grow exponentially. There are only a small number of cases where you really need it,” said Jon Roberts, director of product engineering at Interxion.

CDP makes most sense when applied to mission-critical unstructured file-based information.  Tracking structured-data changes made in the database are best left to the tools of the specific database that is being used.  Two main use cases for CDP are the back up of structured data in ECM or Exchange system, and in protecting desktop and laptop files. CDP is also morphing from being from an alternative type of backup to being just another feature of backup systems that can be selectively applied.

Digg This
Reddit This
Stumble Now!
Buzz This
Vote on DZone
Share on Facebook
Bookmark this on Delicious
Kick It on DotNetKicks.com
Shout it
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter
Google Buzz (aka. Google Reader)

Leave a Reply

Your email address will not be published. Required fields are marked *

*