Research On Network Resource Management Based On Virtual Network Mapping Algorithm

Internet has become one of the most necessary communication mediums in nowadays world, and it has deeply influenced the way of our working and living. However, as the enlarging of the network scale and the rapid increasing of business type, the problems of current Internet facing have appeared gradually, such as scalability, mobility, management, security and credibility. And it is difficult to meet the demand of future network. To solve these problems fundamentally, the researchers at home and abroad have started to design the new framework of future network.Network virtualization is one of the most important research areas of the future network, and is a promising solution that can prevent network ossification. The technology of network virtualization allows multiple heterogeneous networks to cohabit on a shared substrate network, supporting flexible resource management and scheduling, in order to make full use of the resource of substrate physical network. This paper mainly does some research on the technology of resource management and the algorithm of virtual network mapping in the network virtualization.About the technology of resource management in the network virtualization, based on the existing virtual network resource management system, this paper researches a hierarchical and forecasting technology of virtual network resource management. Through predicating the request of network resource, it can allocate the virtual resource dynamically; at the same time, through distributed and intelligent substrate nodes, it can realize the reliability of the virtual resource management.The problem of virtual network mapping can be described as:how to allocate substrate network resource to virtual network requests with respect to their resource requirements, aiming at the efficient usage of physical network resource. In the foundation of existing heuristic algorithm, this paper proposes a new mapping algorithm based on minimum node stress and adjacent principle. During the period of node mapping, the algorithm ensures the balance of node stress, and at the same time using the adjacent principle the link stress is decreased at a certain extent. And the simulation demonstrates that the proposed algorithm reduces the maximum node stress and link stress of the substrate network, and makes it load balance.