云计算环境下基于拓扑感知的虚拟网络映射研究

RESEARCH ON TOPOLOGY AWARE-BASED VIRTUAL NETWORK MAPPING IN CLOUD COMPUTING ENVIRONMENT

网络虚拟技术被认为是克服Internet僵化的一种有效方法,特别是在云计算的环境下。但是虚拟网络映射问题(VNMP)是一个最主要的挑战,其主要是如何通过一种有效的方式将虚拟网络映射到底层网络上从而有效地利用底层的基础资源。虚拟网络映射可以分为两个阶段:节点映射以及链接映射。在节点映射阶段,现有的算法通常使用完全的贪婪策略映射这些虚拟节点,而不考虑这些虚拟节点的拓扑,这将导致底层路径太长(有多个跳跃点)。为解决这一问题,提出一个拓扑感知的节点映射算法,该算法在进行映射时考虑节点的拓扑结构。在链接的映射阶段,新的算法采用k最短路径算法。模拟结果显示,新算法大大增加长期的平均收益,而接受的比率和长期的收益与成本(R/C)成比例。

Network virtualisation technology is recognised as an effective way to overcome the ossification of Internet, especially in the en-vironment of cloud computing.However, virtual network mapping problem （ VNMP） is a major challenge, it focuses on how to map the virtual networks onto substrate network in an effective way so as to efficiently utilise the infrastructure resources in substrate.The problem is also di-vided into two phases：the node mapping phase and the link mapping phase.In node mapping phase, existing algorithms usually map those virtual nodes with complete greedy strategy, without considering the topology among these virtual nodes, and result in the substrate paths be-ing too long （ with multiple hops） .To address this problem, we propose a topology-aware node mapping algorithm, which considers the topol-ogy structure among these virtual nodes.In link mapping phase, the new algorithm will adopt the k-shortest path algorithm.Simulation results show that the new algorithm greatly increases the long-term average revenue, and the acceptance ratio and the long-term revenue to cost ratio （ R/C） are in proportion as well.