Abstract
Data deduplication is a space saving technology that is being used to dramatically improve storage efficiency in the datacenter. This technical session will address the question of what data deduplication is, how it is performed, and the architectural choices available today. The topics covered include source and target deduplication, inline and post-processing, fixed length and variable length segmentation, as well as the availability and integrity of deduplicated data, and the complementary use of replication and removable media. It will also explore the factors affecting space reduction ratios relative to specific deduplication techniques.