Solving a file assignment problem with a genetic plan

Abstract

A genetic plan has been applied to the central server computer network file assignment problem. Throughput of the network is calculated for each file assignment produced by the genetic algorithm. Crossover and mutation are performed on individuals in accordance with their throughput.Various methods of representing and interpreting individuals as file assignments have been attempted. The effect of scaling, and time and space efficiency are addressed. The genetic plan quickly produces file assignments which are near optimal with regard to throughput of the network.