Task Level Partitioning Using the Ant System Algorithm

System design techniques will soon become a necessity in order to tame the immense complexity of future computing systems. This is especially true with the advent of complex system architectures that contain a variety of computing components like microprocessors, memory elements and reconfigurable logic. System partitioning is an important step in the system design process. The problem assigns computational tasks to the different system components. This paper presents a novel approach for the system partitioning problem. Our approach is based on the Ant System algorithm, a meta-heuristic method inspired by the study of the behaviors of ants. In our algorithm, a collection of agents cooperate using distributed and local heuristic information to effectively explore the search space. Experiments show that our algorithm provides robust results that are qualitatively close to the optimal with minor computational cost.