RevDedup

Deduplication is known to effectively eliminate duplicates, yet it introduces fragmentation that degrades read performance. We propose RevDedup, a deduplication system that optimizes reads to the latest backups of virtual machine (VM) images using reverse deduplication. In contrast with conventional deduplication that removes duplicates from new data, RevDedup removes duplicates from old data, thereby shifting fragmentation to old data while keeping the layout of new data as sequential as possible. We evaluate our RevDedup prototype using a 12-week span of real-world VM image snapshots of 160 users. We show that RevDedup achieves high deduplication efficiency, high backup throughput, and high read throughput.