NETWORK CONGESTION ARBITRATION AND SOURCE PROBLEM PREDICTION USING NEURAL NETWORKS.pdfVIP

Smart Engineering System Design, vol. 4:243-252, 2002 NETWORK CONGESTION ARBITRATION AND SOURCE PROBLEM PREDICTION USING NEURAL NETWORKS J. ALAN BIVENS BOLESLAW K. SZYMANSKI Computer Science Computer Science Rensselaer Polytechnic Institute Rensselaer Polytechnic Institute Troy, New York 12 180-3590 Troy, New York 12 180-3590 bivenj @ szymansk@ MARK J. EMBRECHTS Decision Sciences And Engineering Systems Rensselaer Polytechnic Institute Troy, New York 12 180-3590 EMBREM@RPI.EDU ABS TRA CT Rapidly growing needs for networking in the Internet and in intranets make network management increasingly important in todays computer world. We propose using learning techniques to predict network congestion problems before they start impacting the performance of services. In this paper we focus on using a simple feedforward neural network to predict severe congestion in a network. We also use neural networks to predict the source or sources responsible for the congestion, and we design and apply a simple control method for limiting the rate of the offending sources so that congestion can be avoided. Unlike the usual TCP/IP flow control, the proposed method is applied only to selected nodes and converges to the final rate faster. The described techniques set the stage for a new wave of network managers that are capable of preventing networking problems instead of repairing them. KEYWORDS: congestion control, neural networks, weight pruning, network problem prediction, flow rate control INTRODUCTION In the growing world of networking, more and more emphasis is being