Title: Performance improvement of application-specific network on chip using machine learning algorithms

Authors: Mahnaz Rafie; Ahmad Khademzadeh; Midia Reshadi

Addresses: Department of Computer Engineering, Ramhormoz Branch, Islamic Azad University, P.O. Box 151-63825, Tohid Square, Ramhormoz, Iran ' Iran Telecommunication Research Center, P.O. Box 14155-3961, North Karegar Str, Tehran, Iran ' Department of Computer Engineering, Science and Research Branch, Islamic Azad University, P.O. Box 775-14515, End of Sattari Highway, University Square, Martyrs Hisarak Blvd., Tehran, Iran

Abstract: This paper presents a novel and efficient mapping algorithm based on machine learning methods. It produces the best mappings with different metrics which were totally evaluated by support vector machine (SVM), decision tree classifier (DTC), and multiple clustering. It helps to find the optimal application-specific network-on-chip (NoC) based on user's demands on how to customise and prioritise the impact of three key metrics on special mapping. The parameters are robustness index, contention factor and communication cost. In fact, as mapping generator produces a mapping, these parameters will be calculated and compared with some rules. The rules are extracted by SVM and DTC. They highly affect the fitness function of the genetic algorithm (GA). So the algorithm can be controlled. Simulation results show that the proposed algorithm achieves more than fifteen times performance improvement and more supervision in finding the best mappings from numerous generated solutions in comparison with the ordinary algorithm.

Keywords: decision tree; multiple clustering; network on chip; support vector machines; SVM; performance improvement; application-specific NoC; machine learning; mapping algorithm; genetic algorithms.

DOI: 10.1504/IJHPSA.2014.061438

International Journal of High Performance Systems Architecture, 2014 Vol.5 No.2, pp.71 - 83

Received: 01 Dec 2012
Accepted: 28 Aug 2013
Published online: 12 Jul 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article