An Enhanced Node Repeatable Virtual Network Embedding Algorithm Based PSO Solution

摘要：

　　The major challenge in network virtualization is the efficient mapping of virtual nodes and links of virtual networks onto substrate network.In this paper we propose ENR-VNE, an algorithm which can achieves high VN request acceptance ratio in the same time.We modeled VNE problem as an optimal problem to minimize the substrate resource utilization degree.Leverage the advantage of ram data switch between virtual machines host on same physical machine instead of using physical link bandwidth, our algorithm allow repeatable node mapping for same VN.Because the initial value of PSO algorithm is crucial, we present an initial position assign method to accelerate convergence and achieve more repeatable features.Simulation results show that our algorithm achieve high acceptance ratio on same substrate network than un repeatable approach and initial position assign method can further improve the algorithm performance.

关键词： virtual network embedding resource allocation repeatable node mapping particle swarm optimization

作者: Cong Wang Ying Yuan Ying Yang Xi Hu

作者单位: School of Computer and Communication Engineering,Northeastern University at Qinhuangdao, 066004, Chi School of Information Science and Engineering, Northeastern University,Shenyang, 11004, China Liren College of Yanshan University, Qinhuangdao, 066004, China

会议类型: 国际会议

会议名称: 4th international Conference,ICSI2013(第4届群体智能国际会议)

会议地点: 哈尔滨

会议语种:英文

页码: 210-217

在线出版日期: 2013-06-12（万方平台首次上网日期，不代表论文的发表时间）