Title: Evolutionary approach for buffer management in asynchronous transfer mode networks

Authors: M. Sundarambal, R. Devaraj, P. Anbalagan

Addresses: Department of Electrical and Electronics Engineering, Coimbatore Institute of Technology, Coimbatore 641014, Tamil Nadu, India. ' Department of Electrical and Electronics Engineering, Coimbatore Institute of Technology, Coimbatore 641014, Tamil Nadu, India. ' Kalignar Karunanidhi Institute of Technology, Coimbatore 641014, Tamil Nadu, India

Abstract: Fuzzy systems are becoming more widespread in many applications and it is possible to implement using very large scale integration technology. This article deals with buffer management scheme based on evolutionary computing for shared memory asynchronous transfer mode switches. Output queuing is done in the proposed system to increase the throughput. The objective of this research is to model a fuzzy buffer management system that sets the threshold for the output ports in the shared memory. Genetic algorithm technique is used to optimise the fuzzy system parameters derived for each switch size and offers a high degree of scalability. The performance of designed |fuzzy-genetic algorithm| buffer management system is compared with the push-out mechanism (ideal mechanism) and dynamic threshold mechanism. Fuzzy system is simulated using the MODELSIM software and genetic algorithm is simulated using the JAVA language.

Keywords: ATM switches; asynchronous transfer mode switches; buffer management; fuzzy logic; genetic algorithms; information systems; shared memory; VHDL.

DOI: 10.1504/IJISCM.2008.026710

International Journal of Information Systems and Change Management, 2008 Vol.3 No.4, pp.344 - 358

Published online: 23 Jun 2009 *

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