Virtual Network Embedding Based On Content Delivery Network

With the rapid internet development in the half a century or more.As we have used the internet for our daily life,we would say that internet has one of the most important tools that our human being cannot avoid to use it.However,with the rapid development of information technology,it is difficulty for existing internet to meet the requirement of network controllability,scalability,flexibility and quality of service(QoS)and so on,which led to the network virtualization come into being.Network virtualization allows various heterogeneous virtual network(VN)architectures to coexist on a shared substrate infrastructure.Efficient mapping of virtual node and virtual links of virtual network request onto substrate network resources,also known as the virtual network embedding problem,which is the first step toward enabling such large number of users,most of the existing virtual network embedding algorithms mostly focus on utilization of the substrate network,the running time of the algorithm,the acceptance ratio of request and so on,ignoring to the downloading the contents from the content server which sometime lead to be slower,failure,and causing the delay.To tackle the problem,we proposed a way of dealing with VN embedding algorithm which is called VNE-CDN algorithm.The purpose of our algorithm is two aspect.Firstly,we introduce node importance criteria such as degree centrality and closeness centrality measure the average distance and then we will choose the node with the largest bandwidth and CPU resources to map.On the other hand,in order to further accelerate content acquisition,we consider the distance between virtual network and content server.The main task of this paper is to modify and calculate the weights of the substrate nodes by introducing Len weights related to our research objectives in the node mapping stage.For link mapping,virtual links is using k-shortest path algorithm.The simulation results show that the proposed algorithm can significantly reduce the distance between virtual network and content server and accelerate content acquisition without significant loss of acceptance ratio and revenue and cost ratio...