Data deduplication has been gaining more popularity with both clients and vendors across the market. It is a functionality offered by backup, archiving and even storage manufacturers for Tier 2 and Tier 1 storage.
The concept is based on replacing duplicate copies of repeating data blocks with pointers, enabling customers to better manage cost of high-data growth rates. This is "cost avoidance" of expensive disk and tape media.
Deduplication has its place and is exceptionally beneficial, providing a high return on investment when used for the right reasons. But if used inappropriately, it can create more problems than it solves. Certain types of data can be deduped and some cannot. In addition, dedupe ratios will vary according to data types and end users should be cautious of figures indicated by vendor marketing literature.
Deduplication can be performed at different levels of the storage tiers or even at the host level. For Backup, global deduplication can also be used to remove redundant copies of data across the Wide Area Network, delivering cost saving benefits of reduced management, disk, tape media and network bandwidth. But it is also important to remember that accessing, backing up or recovering deduped data will need to be re-hydrated to its original size. Hence capacity planning is critical.
Servnet are ideally placed to help customers understand how deduplication works at different levels, and where it will offer the greatest benefit to an individual organisation, based on their challenges, environment and data type. We also conduct sizing exercises for deduplication and design workshops right through to implementation to ensure the customer achieves the right solution.