Yiying Zhang and Vijayan Prabhakaran
We propose a duplication-aware disk array (DADA), which instead of removing duplicates, keeps track of block duplication and uses duplicate contents to improve the reliability and availability of storage arrays. DADA is designed specifically for primary storage systems, which contain only a moderate degree of duplication and therefore, do not benefit from deduplication. Based on our analysis of 5 different file server contents, we show that DADA can reduce scrubbing and recovery time by 17-26% with little or no overhead. DADA can also use its duplication awareness to recover from latent sector errors that are otherwise unrecoverable.
Publisher Microsoft Technical Report