Virtual network embedding algorithm based on a hybrid swarm intelligence optimization
Wang Wenzha · Journal of Computer Applications · 2014
Network virtualization is recognized as a significant technology to solve the ossification of current Internet. Virtual Network Embedding( VNE) is a major challenge in network virtualization. The main object of VNE is to increase the acceptance ratio of Virtual Network( VN) and the revenue of infrastructure providers. Regarding VNE as an Integer Linear Programming( ILP) model with an assumption that substrate network needs to support path splitting, a new VNE algorithm based on hybrid swarm intelligence optimization was proposed. The proposed algorithm took advantage of the Genetic Algorithm( GA) and Particle Swarm Optimization( PSO) to optimize the mapping scheme in view of the balance of the mapping overhead and the mapping proportionality. Compared with the existing mainstream approaches, the simulation results demonstrate that the proposed algorithm can increase the long-term average revenue and acceptance ratio.