Dynamic Replication Management Scheme for Distributed File System

Nowadays, replication technique is widely used in data centerstorage systems to prevent data loss. Data popularity is a key factor in datareplication as popular files are accessed most frequently and then they becomeunstable and unpredictable. Moreover, replicas placement is one of key issuesthat affect the performance of the system such as load balancing, data localityetc. Data locality is a fundamental problem to data-parallel applications thatoften happens and this problem leads to the decrease in performance. To addressthese challenges, this paper proposes a dynamic replication management schemebased on data popularity and data locality; it includes replica allocation andreplica placement algorithms. Data locality, disk bandwidth, CPU processingspeed and storage utilization are considered in the proposed data placementalgorithm in order to achieve better data locality and load balancing effectively.Our proposed scheme will be effective for large-scale cloud storage