Multiple tasks allocation in arbitrarily connected distributed computing systems using A* algorithm and genetic algorithm

Abstract

A number of algorithms is proposed for allocation of tasks in a DCS. Most of them did not consider allocation of various unrelated tasks partitioned into modules by taking into account the architectural capability of the processing nodes and the connectivity among them. This work considers allocation of disjoint multiple tasks with corresponding modules wherein multiple disjoint tasks with their modules compete for execution on an arbitrarily networked DCS. Two algorithms have been presented based on well-known A*algorithm and Genetic Algorithm techniques. The proposed algorithms consider a load balanced allocation for the purpose. The paper justifies the effectiveness of the proposed algorithms using several case studies. © Springer-Verlag 2006.