Article


Article Code : 139308051041292809(DOI : 10.7508/jist.2015.02.004)

Article Title : Ant Colony Scheduling for Network On Chip

Journal Number : 10 Spring 2015

Visited : 912

Files : 436 KB


List of Authors

  Full Name Email Grade Degree Corresponding Author
1 Neda Dousttalab n.dousttalab@gmail.com - -
2 Mohammad Ali Jabraeil Jamali m_jamali@itrc.ac.ir - PhD
3 Ali Ghaffari a.ghaffari@iaut.ac.ir Associate Professor PhD

Abstract

The operation scheduling problem in network on chip is NP-hard; therefore effective heuristic methods are needful to provide modal solutions. This paper introduces ant colony scheduling, a simple and effective method to increase allocator matching efficiency and hence network performance, particularly suited to networks with complex topology and asymmetric traffic patterns. Proposed algorithm has been studied in torus and flattened-butterfly topologies with multiple types of traffic pattern. Evaluation results show that this algorithm in many causes has showed positive effects on reducing network delays and increased chip performance in comparison with other algorithms.