Survivable virtual network embedding algorithm considering multiple node failure in IIoT environment

Abstract

The Industrial Internet of Things is a new ecology formed by the deep integration of the Internet of Things and the industrial environment. Network is one of the core technologies of industrial Internet of things. However, in the face of billions of industrial Internet of Things applications, the traditional network architecture cannot meet the needs of these applications for network resources. Network virtualization technology can effectively solve the above problems by decoupling network services from the underlying network. However, the existing heuristic virtual network embedding algorithm is easy to fall into the local optimum, such as the virtual network embedding algorithm based on particle swarm optimization. Therefore, in order to solve the above problems, a virtual network embedding strategy based on the hybrid whale optimization algorithm is proposed. This strategy can avoid the introduction of the population into the local optimum while reducing the amount of calculation. Secondly, a node importance measurement strategy based on weighted aggregation coefficient and traffic prediction is introduced. This strategy can conduct a more reasonable and effective assessment of the importance of nodes. Thirdly, to reduce resource occupation and improve network security and stability, we propose a multi-node failure recovery strategy. It depends on the construction of candidate nodes and the selection of recovery points. Simulation results show that compared with other algorithms, the proposed algorithm reduces the cost and maintains a high success rate of fault recovery. In addition, its request acceptance rate is twice that of other algorithms.