Genetic list scheduling for soft real-time parallel applications

Abstract

This paper presents a hybrid algorithm that combines list scheduling with a genetic algorithm for constructing nonpreemptive schedules for soft real-time parallel applications represented as directed acyclic graphs. The execution time requirements of the applications' tasks are assumed to be stochastic and are represented as probability distribution functions. The approach presented here produces shorter schedules than two popular list scheduling approaches for a majority of sample problems. Furthermore, the stochastic schedules provide a mechanism for predicting the probability of the application completing when the execution time available is less than the worst case requirement.