作者: Yijie Wang , Sijun Li
DOI: 10.1016/J.CAMWA.2006.05.002
关键词:
摘要: Replication is a key technology of distributed storage systems. In this paper, an indirect replication algorithm is proposed following the intrinsic characteristic of distributed storage systems and the peer-to-peer model. In the indirect replication algorithm, the data object is partitioned into several data blocks. These data blocks are encoded in order that there is data redundancy between data blocks. Comparing with the traditional replication algorithm, the indirect replication algorithm has less granularity of replication, less bandwidth and …