一种拓扑感知的虚拟网络重构算法  

A Topology-awareness Virtual Network Reconfiguration Algorithm

作　　者：彭利民[1] 

机构地区：[1]华南农业大学数学与信息学院,广东广州510642

出　　处：《四川大学学报（工程科学版）》2015年第5期110-115,共6页Journal of Sichuan University (Engineering Science Edition)

基　　金：国家自然科学基金资助项目(61103037);广东省自然科学基金资助项目(S2012040007599)

摘　　要：针对底层物理网络负载不均衡问题,根据节点和链路的负载状态,动态地选择负载过大的物理节点和物理链路进行重构。通过建立拓扑感知的虚拟网络重构模型,将虚拟节点迁移到距离其邻居映射节点最近的物理节点上,优化底层物理网络中虚拟重构节点的映射位置,降低虚拟网络重构的资源开销。模拟结果表明,提出的TAVNR算法有效地均衡了底层物理网络的负载水平,降低了虚拟链路的重构路径长度,提高了虚拟网络请求接受率。Aiming at load imbalance problem in the substrate network,the substrate nodes and links with resource utilization exceed the predefined threshold value were selected for reconfiguration based on the load state of the substrate nodes and links. By using topology- awareness reconfiguration model, the virtual node to be migrated was then mapped onto the nearest mapped substrate node, which is the neighbor mapping node of the migration virtual node, mapping location of these reconfiguration virtual nodes in the substrate network was optimized, therefore resource consumption of virtual network reeonfiguration was reduced efficiently. Simulation results showed that the load level in the substrate is balanced efficiently, the reconfiguration path distance of virtual links are reduced greatly, and the accept ra- tio of virtual networks is improved by using the proposed TA-VNR algorithm.

关 键 词：底层物理网络 虚拟网络 重构 资源利用率 拓扑感知 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]