Genetic Algorithms versus Tabu Search for Instruction Scheduling

Abstract

Most scheduling problems require either exponential time or space to generate an optimal answer [7]. Instruction scheduling is an instance of a general scheduling problem and Dewitt [8] uses this fact to show instruction schedul- ing is a NP-complete problem. This paper applies Ge- netic Algorithms, Tabu Search, and list scheduling to the instruction scheduling problem and compares the results obtained by each.