Complete guide to backup deduplication
A comprehensive collection of articles, videos and more, hand-picked by our editors
Semantic deduplication, more commonly known as semantic-aware multi-tiered deduplication, or SAM, is a legacy deduplication process that has existed since at least 2010.
By submitting your personal information, you agree that TechTarget and its partners may contact you regarding relevant content, products and special offers.
The technology was supposed to strike a balance between deduplication ratios and the overhead required by the deduplication process. When semantic deduplication was introduced, achieving high deduplication ratios meant committing significant hardware resources to the deduplication process, which tended to impact performance. More performance-friendly deduplication algorithms of the time were less effective with regard to the level of deduplication they could achieve.
Striking a balance between deduplication and performance is much less of an issue than it was five years ago. Deduplication is now a mature technology, and much of the overhead produced by the deduplication process can be countered by more efficient deduplication algorithms and by hardware offloading.
Semantic deduplication worked by taking a multi-tier approach to the deduplication process. File data was globally deduplicated in an effort to remove redundant files. Data was also deduplicated locally at the chunk level. This multi-tier approach resulted in higher deduplication ratios with less overhead.
Even though semantic deduplication is a legacy technology, there are a number of products in use today that perform global deduplication using a very similar technique. Imagine that 10 servers need to be backed up, and each server runs the same operating system. Performing a block-level deduplication on the servers would eliminate redundancy at the server level, but redundancy would still exist within the backup target due to similarities in the various servers. A second deduplication pass at the backup target level can eliminate any cross-server redundancy. While this approach isn't exactly the same as semantic deduplication, the similarities cannot be ignored.
Available deduplication options
Vendors trying to speed up the deduplication process
Deduplication can benefit your organization
Related Q&A from Brien Posey
Setting up Office 365 generally involves multiple devices. With nonpersistent VDI, the rules of the game change for IT admins.continue reading
Much has been said about the inability to scale storage separately from other resources in a hyper-converged system, but are there any advantages to ...continue reading
The definition of hyper-converged infrastructure has evolved as the technology has grown. But the phrase still means different things depending on ...continue reading
Have a question for an expert?
Please add a title for your question
Get answers from a TechTarget expert on whatever's puzzling you.