Weight Based Prioritization of Replica for Replica Replacement in Data Grid

摘要

Datagrid provides storage facility to data grid user to store and process huge amount of data. Replication plays a major role in optimizing the Datagrid. The replacement of replica in datagrid is to be optimal to improve the performance of the Datagrid. The replacement of replica is decided based on the priority of the replica in future. The priority of the replica is based on its importance in the near future. In this paper, an approach is proposed to evaluate the priority of the replica by considering the importance of that replica in the near future. The proposed approach is evaluating the priority using the rubrics on the parameters; Frequency of access (F), Number of Access (N), and Recent access (R). The rubrics on each parameter are constructed empirically. The priority of the replica (P_R) is determined by using the rubrics values and weights. The proposed approach is simulated in the OptorSim simulator. The simulation result shows that the proposed approach yielded a better Hit Rate consistently on various file sizes.