计算机工程与应用 ›› 2015, Vol. 51 ›› Issue (23): 94-99.

• 网络、通信、安全 • 上一篇    下一篇

基于熵权法的虚拟网络映射算法

许  倩1,2,3,易辉跃2,3,朱  军1,胡宏林2,3   

  1. 1.安徽大学 电子信息工程学院,合肥 230039
    2.上海无线通信研究中心,上海 200335
    3.中国科学院 无线传感网与通信重点实验室,上海 200335
  • 出版日期:2015-12-01 发布日期:2015-12-14

Virtual network mapping algorithm based on entropy weight method

XU Qian1,2,3, YI Huiyue2,3, ZHU Jun1, HU Honglin2,3   

  1. 1.School of Electronics and Information Engineering, Anhui University, Hefei 230039, China
    2.Shanghai Research Center for Wireless Communications, Shanghai 200335, China
    3.Key?Laboratory?of?Wireless?Sensor?Network?&?Communication, Shanghai?Institute?of?Microsystem?and?Information?Technology, Chinese?Academy?of?Sciences, Shanghai 200335, China
  • Online:2015-12-01 Published:2015-12-14

摘要: 目前虚拟网络研究的一个热点是虚拟网络映射,但是传统两阶段算法中的节点映射算法着重于提高网络资源利用率,而忽略了网络的整体负载性能。为了避免现有映射算法中使用单一固有属性计算拓扑势值带来的片面性,在节点映射过程中增加了节点的另一个固有属性。但是,由于这两个固有属性之间的数量级相差较大,从而引入熵权,通过计算两个属性的熵权值来优化拓扑势值的计算,提出了一种基于熵权法的虚拟网映射算法。仿真实验结果表明,所提出的算法提高了映射接受率,并降低了网络的整体负载。

关键词: 网络虚拟化, 虚拟网络映射, 拓扑势, 熵权

Abstract: Virtual network mapping in network virtualization has been attracted much attention in recent years. Most node mapping algorithms focus on the utilization efficiency of the network resources, but ignore the performance regarding to the load balance of the whole network. In order to avoid one-sideness when it only uses one intrinsic attribute to compute the topology potential value in the existing mapping algorithm, this paper adds another node?inherent?attribute?in?the process of the node mapping. Because the magnitude of these two inherent properties are not in the same order, it introduces the entropy weight, and optimizes the calculation of the topology potential value through computing the entropy value of these two attributes. Then, this paper proposes a virtual network mapping algorithm based on the entropy weight method. Simulation results show that the proposed algorithm improves the success rate of the mapping and reduces the network overall load.

Key words: network virtualization, virtual network mapping, topology potential, entropy weight