Heterogeneous Data Storage Management with Deduplication Aware in Cloud Computing

Abstract

Cloud computing greatly facilitates information suppliers who got to be compelled to provide their information to the cloud whereas not revealing their sensitive information to external parties and would love users with sure credentials to be able to access the info. knowledge reduction has become further and really vital in storage systems as a results of the explosive growth of digital info among the world that has ushered among the large information era. one altogether the foremost challenges facing large-scale info reduction may even be a due to maximally sight and Elimination theme that effectively exploits existing duplicate-adjacency information for terribly economical likeness detection in information deduplication based absolutely backup/archiving storage systems. the foremost created behind DARE is to use a problem, decision Duplicate closeness based absolutely likeness Detection (DupAdj), by considering any two information chunks to be similar Duplicate closeness primarily based utterly likeness Detection (DupAdj), by considering any 2 data chunks to be similar if their individual adjacent info chunks square measure duplicate throughout a} terribly deduplication system, then associatey enhance the likeness detection efficiency by associate improved super-feature approach. Our experimental results supported real-world and artificial backup datasets show that DARE entirely consumes concerning 1/4 and 1/2 severally of the computation and assortment overheads required by the standard super feature approaches whereas detection 2-10% extra redundancy and achieving succeeding output, by exploiting existing duplicate closeness knowledge for likeness detection and finding the sweet spot for the super feature approach.