科研之友
微信
新浪微博
Facebook
分享链接
摘要
As one of the challenges for network virtualization, virtual network embedding which maps Virtual network (VN) to the substrate network and allocates resources according to the requirements of VN in an efficient way has gained great attention. Existing algorithms generally make their decision according to the present available substrate network resource, especially bandwidth. This paper proposes a time-based VN embedding algorithm. A probability model is formulated to obtain the maximum probability that the available resources of substrate network can be used by succeeding VN requests. The probability is set as the weight of the node and the greedy algorithm is employed to embed the virtual node. The reciprocal of the probability is set as the weight of the link and the shortest path algorithm is employed to embed the virtual link. Simulation experiments show that the proposed algorithm increases the acceptance rate and the revenue compared to the existing algorithms.
