A load balancing algorithm based on key-link and resources contribution degree for virtual optical networks mapping

Abstract

Aiming at efficiently mapping multiple virtual optical networks (VONs) on a shared substrate network. This paper proposes a load balancing algorithm that is based on key-link and resources contribution degree (RCD) for VONs mapping (KLRVM) in elastic optical networks (EONs). In this algorithm, virtual nodes mapping gets the network status according to the proposed RCD, and comprehensively considers nodes computing resource, shortest path length and available bandwidth. While the link length, the occupied link bandwidth and key-link parameters are holistically considered to complete the virtual links mapping, thus bottle-neck links of a network could be avoided. Simulation results show that the proposed algorithm can effectively improve the balance performance of network load and reduce the spectrum resource usage.