Adaptive Granularity Algorithm for Effective Distributed Load Balancing and Implementation in Multiprotocol Label Switching Networks

Abstract

Multiprotocol label switching (MPLS) is an extension to the existing Internet protocol (IP) architecture. Explicit routing of MPLS facilitates balancing the load by splitting traffic and moving from a congested part of the network to some other part in a well controlled way. To avoid harmful oscillations, splitting ratios are varied only with finer granularity steps. However, finer granularity increases the number of iterations to reach the steady state optimal solutions. To overcome this problem we are proposing an adaptive granularity algorithm for MPLS networks. A numerical method is developed to evaluate the performance of the algorithm. The results show that the link utilization and the mean delay obtained in a reasonable number of iterations are very close to the optimal values and the throughput is improved in MPLS network. The network simulator ns2 version ns2.31 is used for simulation.