A Load Balancing Algorithm Based on End-to-end Rerouting

Abstract

With the quick growth of new Internet applications, the network load is increasing rapidly which may cause network congestion and has a certain impact on the transmission performance and user experience. There are two ways to solve this problem: increasing network bandwidth and balancing network load to increase network utilization. This paper mainly discusses the latter. As we all know, the shortest path algorithm is common for network routing, but it cannot balance the load between the network links, causing the load on some links heavy, we call the link with the heaviest load as the peak link. This paper proposes a new Load Balancing algorithm based on End-to-end Rerouting (LBER). First, the peak link is obtained according to the load status of the network links. Then, the load of the peak link is decreased by additional cost to punish it, that all the paths through the peak link are rerouted from end to end. The ultimate goal is to minimize the peak link load. Simulation experiments show that this algorithm can effectively reduce the peak traffic of the network compared with the existing shortest path algorithm by an average of 29.29%, which means the network bandwidth utilization is improved notably.