Virtual Network Embedding Based on Modified Genetic Algorithm
Peiying Zhang,Haipeng Yao,Maozhen Li,Yunjie Liu
DOI: https://doi.org/10.1007/s12083-017-0609-x
IF: 3.488
2017-01-01
Peer-to-Peer Networking and Applications
Abstract:To fend off the ossification of Internet architecture, virtual network embedding has been propounded as one of the most important techniques to address this issue. Virtual network embedding is a process that consists of two stages including node mapping stage and link mapping stage, the aim of node mapping stage is to map the virtual nodes from virtual network requests (VNRs) onto the substrate nodes meanwhile satisfying the CPU capacity constraints on nodes, the goal of link mapping stage is to map the virtual links from VNRs onto the substrate paths while satisfying the bandwidth resource constraints on links. This paper proposed a virtual network embedding algorithm based on modified genetic algorithm, improved the classical genetic algorithm from three aspects: population initialization strategy, improved mutation operation and improvement operation, took advantage of the selection operation, crossover operation, mutation operation, feasibility checking operation, and utilized the fitness function to choose the best chromosome. Simulation results indicated that our proposed method has significantly increased the acceptance ratio of VNRs and the long-term average revenue of Infrastructures (InPs) compared with other two state-of-the-art algorithms.