A DPSO-Based Load Balancing Virtual Network Embedding Algorithm with Particle Initialization Strategy

Abstract

Virtual network embedding (VNE) is helpful for effective sharing the underlying physical substrate network resources. In the existing particle swarm optimization-based VNE algorithms, the particle position initialization phase will produce a lot of resources fragment and prevent the underlying substrate network from receiving more resource requests. In order to improve the utilization rate and load balancing of the substrate network as well as accelerate the mapping efficiency of VNE algorithm, this paper presents a particle swarm optimization based load balancing VNE algorithm. A particle position initialization strategy is introduced by reducing the candidate substrate nodes to reject the inaccuracy caused by the random selection in position initialization. Simulation results show that the proposed algorithm can effectively increase the acceptance rate of virtual network requests and improve the long-term average revenue to cost ratio of the substrate network.