Dynamic Task Scheduling in Multiprocessor Real Time Systems Using Genetic Algorithms

Abstract

The objective of the scheduling algorithm is to dynamically schedule as many tasks as possible such that each task meets its execution deadline while minimizing the total delay time of all of the tasks. The problem of scheduling of real-time tasks in multiprocessor systems is to determine when and on which processor a given task executes. In this paper we suggest a genetic algorithm for dynamic scheduling of real time tasks in multiprocessors system.The algorithm based on the use of a fixed size chromosome and repeatedly applying specific crossover (single point or double point) and mutation procedures with variable mutation rates (0.05 – 0.1) until all tasks are successfully scheduled.