Heterogeneous Processors Scheduling Problems using MAX-MIN Ant System & Crossover Procedure

Abstract Recently, the advances of the Internet concept and the existence of the high-speed networks as low-cost commodity components are changing the way we use popular computers today and active research area has given the emergence of a new paradigm identified by the concept of grid computing system and the Heterogeneous Processors Scheduling Problem (HPSP). The HPSP is a fundamental step for mapping a set of jobs to computational device processors. The main objective is to minimize the completion time noted makespan of the given HPSP while effectively using the computational resource processors. The type of scheduling problem is NP-hard, thus effective heuristic methods are necessary to provide a qualitative scheduling solution. In this paper, we introduce an extension of the Ant System metaheuristic that combines the MAX-MIN Ant System and Crossover Procedure concept. Indeed, at the saturation of the pheromones trail the Crossover Procedure is applied automatically to diversify the space research and improve the Ant System solution and to update the pheromone trail. In fact , we propose a hybrid metaheuristics for the Heterogeneous Processors Scheduling Problem. A 32-Jobs/4-Processors example shows the effectiveness of the developed method.