This paper presents a set of techniques to substantially reduce disk I/Os in high-throughput deduplication storage systems. Our experiments show that the combination of these techniques can achieve over 210 MB/sec for 4 multiple write data streams and over 140 MB/sec for 4 read data streams on storage server with two dual-core processors and one shelf of 15 drives.

These techniques are general methods to improve throughput performance of deduplication storage systems. Our techniques for minimizing disk I/Os to achieve good deduplication performance match well against the industry trend of building many-core processors.