Solving the File Allocation Problem in the Distributed

Abstract

Average Distributed Program Throughput (ADPT) of the Distributed Computing System (DCS) depends mainly on the allocation of various resources. One of the important resources to be allocated on a DCS is various files. In this paper, we propose an approach that uses genetic algorithms to determine the optimal file allocation on the DCS that maximizes the ADPT with the constraint that the total number of copies of each file on a DCS must be equal to or less than the specified value. The algorithm has been applied on different network examples taken from literature; the results show that the algorithm is efficient to obtain better results