Summary
The hype around Data Domain that brought it $2.1 Billion, over-emphasized the role of deduplication. The technology is nice, but it is neither unique nor complete.
Analysis
Deduplication is a very nice technology. However, it is only a small piece of the backup problem.
Data Backup has not changed much in the past 40+ years. It still revolves around the notion of a series of sequential magnetic tapes, or more recently virtualized tapes on disk storage. Deduplication alone, make this archaic process a bit less cumbersome, but does not fundamentally change or solve the bigger issues with this strategy.
Traditional Data backup has several problems, including cost, security, processing time, and more importantly restoration capabilities (AKA eDiscovery). These strategies also do not provide for automatic destruction of historic data at the specified time horizon.
A more modern paradigm would look at backups more like archived data. Using modern technology like deduplication, large lower cost storage arrays, and indexing techniques would create a more cost effective, automated, and safer solution. Companies that are developing these solutions like Symantec and Commvault already have deduplication capabilities. I believe that the future of backups/archiving lies with holistic solutions that redefine the strategy, not simply making the old one more efficient.
Howard Bruck consults with leading institutions through GLG
Analyses are solely the work of the authors and have not been edited or endorsed by GLG.
