DI_GA: A Heuristic Mapping Algorithm for Heterogeneous Network-on-Chip

Abstract

Heterogeneous multi-core processors have become the forefront of processor development due to their advantages in system throughput and execution efficiency, but they also bring many new challenges to system design. The mapping of heterogeneous Network-on-Chip (NoC) is challenging. To solve this problem, this paper proposes a heterogeneous multi-core processor task mapping algorithm based on an improved genetic algorithm. By constructing a good initial population method to improve the initial population quality, a dual population genetic mechanism is used in the iteration process. The algorithm can make tasks more reasonably distributed to various network nodes, and has high efficiency for optimizing network power consumption on heterogeneous multi-cores.