文章基本信息

标题：DELTA REFERENCE DATA DEDUPLICATION IN LOW UPLOAD BANDWIDTH NETWORK
本地全文：下载
作者：XIANZHUO LIU ; JINLIN WANG ; MING ZHU 等
期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2013
卷号：53
期号：1
出版社：Journal of Theoretical and Applied
摘要：When users put their files on remote storage, it is important to update local changes in short time. Due to the redundancy between successive versions of files, compare-by-hash and delta compression have been studied to reduce total volume locally transmit. Better performance are desirable since plain compare-by-hash does not fully exploit redundancy and delta compression lay additional space demands and application oriented limits on local storage system. In this paper, a delta reference approach is proposed to further reduce the volume sent to remote server by dynamically searching for similar file and reference chunks for the chunk sent from the local host. Experiments on practical datasets reveal that the proposed approach can reduce the volume sent to remote server up to 28.3% thereby decreasing transfer time as much as 26.2% in typical low upload bandwidth networks.
关键词：Delta Compression; Data Deduplication; Content Defined Chunking; Bandwidth Asymmetric Network; Data Synchronization