基于节点邻近感知与路径综合评估的虚拟网络映射算法

发布时间：2018-11-27 10:04

【摘要】：该文针对虚拟网络映射过程中节点与链路映射关联性差、虚拟邻接节点映射距离远和相邻节点链路资源消耗不均衡的问题,提出一种两阶段的基于节点邻近感知与路径综合评估的虚拟网络映射算法(NA-PVNM)。在节点映射阶段,首先对虚拟节点按照资源请求和广度优先搜索算法排序,然后对每个虚拟节点的候选物理节点考虑资源丰富度和拓扑邻近性,建立了节点适应值函数计算最佳映射节点。在链路映射阶段,采用k最短路径算法,对每条候选路径考虑可用带宽、途经物理节点资源最大值和路径跳数,通过路径适应值函数计算最佳路径。仿真结果表明算法降低了链路映射长度,提高了虚拟网络请求接受率和收益开销比。实验分析了位置约束和物理网络拓扑性质对算法性能的影响以及映射过程中物理网络资源占用情况。实验结果表明,在物理资源分布和虚拟网络请求的条件约束下,提高虚拟网络映射成功率的关键在于减少映射时的资源消耗。
[Abstract]:This paper aims at the problems of poor correlation between nodes and link mapping, long mapping distance of virtual adjacent nodes and unbalanced consumption of link resources of adjacent nodes in the process of virtual network mapping. A two-stage virtual network mapping algorithm (NA-PVNM) based on node proximity perception and path evaluation is proposed. In the node mapping phase, the virtual nodes are sorted according to resource request and breadth-first search algorithm, and then resource richness and topology proximity are considered for each virtual node candidate physical node. The node fitness function is established to calculate the optimal mapping node. In the phase of link mapping, k shortest path algorithm is used to calculate the best path by considering the available bandwidth for each candidate path, passing through the maximum resource and the number of hops of the physical node, and calculating the best path by the path fitness function. Simulation results show that the algorithm reduces the length of link mapping and improves the request acceptance rate and the ratio of revenue to overhead of virtual network. The effects of location constraints and topological properties of physical networks on the performance of the algorithm and the resource occupancy of physical networks in the mapping process are analyzed experimentally. The experimental results show that under the constraints of physical resource distribution and virtual network request, the key to improve the success rate of virtual network mapping is to reduce the resource consumption.
【作者单位】：空军工程大学信息与导航学院;
【基金】：国家自然科学基金(61401499)~~
【分类号】：TP393.01

【参考文献】