Accelerating Content-Defined Chunking for Data Deduplication Based on Speculative Jump

In data deduplication systems, chunking has a significant impact on the ratio and throughput. Existing Content-Defined Chunking (CDC) approaches exploit sliding window to calculate rolling hashes of input stream byte-by-byte, then determine chunk cut-points if hash satisfies given cut-condition. Since previous CDC are extremely costly, it often significantly degrades throughput systems. this paper, we argue that calculating checking byte-by-byte is unnecessary. To reduce CPU overhead CDC, propose xmlns:xlink="http://www.w3.org/1999/xlink">jump-based chunking (JC) approach. The key idea introduce jump-condition, can jump over specific length satisfy jump-condition. Moreover, also explore cut-condition jump-condition size. Our theoretic studies demonstrate effectiveness efficiency JC, without compromising ratio. Experimental results show JC improves by about 2× average compared with state-of-the-art while still guaranteeing high