A Utility-Based Data Replication Algorithm in LargeScale Data Grids

Data grids support access to widely distributed storage for large numbers ofusers accessing potentially many files. To enhance access time, replication atnearby sites may be used. Data replication, a technique much investigated by datagrid researchers in past years creates multiple replicas of file and places them inconventional locations to shorten file access times. One of the problems in datareplication is creation of replicas, replica placement and replica selection.Dynamic creation of replicas in an appropriate site by data replication strategycan increase the systems performance. In this paper, we propose a data replicationalgorithm, called the Utility-base Data Replication (UDR) algorithm that improvesfile access time. Each grid site has its own capabilities and characteristics;therefore, choosing one specific site from many sites that have the needed data is akey and significant decision. The replica selection problem has been studied bymany researchers who only considered response time as a criterion for the selectionprocess. Therefore, in this study, we addressed the problem of how to select the bestreplica for the users jobs. Our approach is simulated using a data grid simulator,OptorSim, developed by European Data Grid projects. Comparing to the previouswork the experimentation shows the improvement in the overall performance.