Gossiping for autonomic estimation of network-based parameters in dynamic environments

Abstract

Future networks are becoming larger in scale, more dynamic and heterogeneous. In such environments, adaptation of algorithms in the current networking conditions is necessary in order to increase the efficiency and performance of the deployed mechanisms. Knowledge of network based parameters may facilitate the decision making processes in a dynamic environment. Self-awareness will help the network to diagnose faults and realize the current status and, thus, proceed to self-optimization actions. In this paper, a neighbour to neighbour gossiping mechanism is proposed for autonomic estimation of network based parameters. The described mechanism is evaluated in terms of message exchanges and convergence time until accurate parameters' estimation. © 2010 Springer-Verlag Berlin Heidelberg.