Robust Inline Data Reduction Technique in Multi-tenant Storage

Data deduplication has gained increasing popularity as a space-reduction approach in backup storage systems. One of the main challenges for centralized data deduplication is the scalability of fingerprintindex search. In existing system, deduplication mainly focuses on backup system. In this paper, we propose a system that effectively exploits similarity and locality of data blocks to achieve high duplicate elimination and well balanced load at low RAM overhead in a runtime storage environment. The main idea is to employ an algorithm by considering random data chunks which are similar with adjacent data chunk, and eliminate duplicate chunks from the storage system, and then further enhance the resemblance detection efficiency. This would further help us to enhance the data delivery capabilities in cloud storage system with reduced latency and high throughput.