Smart Vogels Approximation Method SVAM

Data Grid technology is designed to handle large-scale data management for worldwide distribution, primarily to improve data access and transfer performance. Several strategies have been used to exploit rate differences among various client-replica provider links and to address dynamic rate fluctuations by dividing replicas into multiple blocks of equal sizes. However, a major obstacle, the idle time of faster providers having to wait for the slowest provider to deliver the final block, makes it important to reduce differences in finishing time among replica servers. In this paper, we propose a dynamic optimization method, namely Smart Vogel's Approximation Method, to improve the performance of data transfer in Data Grids. Our approach reduces the differences that ideal time spent waiting for the slowest replica provider to be equal or less to the predefined data transfer completion time with minimum prices of replicas.