A Prediction-Based Replication Algorithm for Improving Data Availability in Frid Environment

Abstract

Data replication is a key optimization technique for reducing access latency and managing large data by storing replica of data in a wisely manner. In this paper, we propose a data replication algorithm, called the Prediction-Base Dynamic Replication (PBDR) algorithm that improves file access time. Restricted by the storage capacity, it is essential to design an effective strategy for the replication replacement task. PBDR deletes files by considering four important factors: the number of requests for the replica in the future times, availability, the size of the replica and the last time the replica was requested. Also, it can minimize access latency by selecting the best replica when various sites hold replicas of datasets. The algorithm is simulated using a data grid simulator, OptorSim, developed by European Data Grid projects. The experiment results show that PBDR strategy gives better performance compared to the other algorithms and prevents unnecessary creation of replica which leads to efficient storage usage.