论文部分内容阅读
研究目的:基于虚拟网络请求和底层物理网络实时拓扑属性,提出一种高效的两步式虚拟网络映射算法。创新要点:分别利用中介中心性和物理节点相关性对虚拟网络请求和底层物理网络中节点进行重要性评估,在此基础上给出一种两步式映射算法(算法1,2)。研究方法:首先给出中间中心性、接近中心性以及节点相关性计算模型,结合节点本地资源分别提出虚拟网络请求和物理网络中节点排名计算方式。当虚拟网络请求到达后,根据虚拟节点排名,将其映射到拥有足够资源的物理节点中排名最靠前的节点。节点映射完成后,使用K-th最短路径算法进行链路映射。映射过程中采用文献(Yu et al.,2008)中所使用的时间窗口模式进行接入控制。重要结论:利用节点本地资源,针对性分析虚拟网络请求和物理网络实时拓扑属性,提出两步式映射算法。该算法提高请求接受率、开销收益比的同时减少算法映射时间,取得更好的映射效果(图3-10)。
Research purposes: Based on the virtual network request and the real-time topology properties of the underlying physical network, an efficient two-step virtual network mapping algorithm is proposed. Innovative Points: We evaluate the importance of virtual network requests and nodes in the underlying physical network by using the relativity between the centrality and the physical nodes, respectively. Based on this, we present a two-step mapping algorithm (Algorithm 1 and Algorithm 2). Research methods: Firstly, we give a calculation model of centrality, nearness and node correlation, and put forward virtual network request and node ranking calculation method in physical network with local resources of nodes. After the virtual network request arrives, it is mapped to the top-ranked node among the physical nodes that have sufficient resources according to the virtual node rankings. After the node mapping is completed, the link mapping is performed using the K-th shortest path algorithm. The mapping process uses the time window model used in the literature (Yu et al., 2008) for access control. Important conclusions: With the local resources of nodes, the virtual network request and the real-time topology properties of physical network are analyzed in detail, and a two-step mapping algorithm is proposed. The algorithm improves the request acceptance rate, overhead and revenue ratio while reducing the algorithm mapping time, get a better mapping effect (Figure 3-10).