A Dynamic Convergent Replica Selection Strategy Based on Cloud Storage

Replica selection strategy have been used in Cloud Storage to reduce data access latency, eliminate bottlenecks of hot spots and achieve load balance. This paper studies the replica selection strategy and replication strategy. By concerning real-time bandwidth, performance of the peer, and replica's history request, following the principle that high valued replica has high proportionality to be select, a proportional matching based self-convergent replica selection strategy (PMC) based on HDFS (Hadoop Distributed File System) cloud storage environment is presented. The strategy has global optimization, effectively reduces access delay, balances load, eliminates hot bottlenecks. The simulation results show that the optimal selection strategy offers better performance by shortening the response time of replica selection.