Scheduling of resources for a grid computing application

Abstract

This research work considers a job that consists of three tasks (face recognition, voice recognition and finger print) each that has been submitted for schedule over a set of resources (computer processor) with an objective to minimize tardiness. Genetic Algorithm was used to solve this problem. While obtaining an optimal solution is not always guaranteed, results show that a ‘satisficing’ solution can always be obtained. It was discovered that increase in iteration provides a higher chance for a better solution with tardiness minimized. Obtaining a quality solution, with increase in number of jobs, is proportional to the number of generations in the scheduling program.