An Improved Genetic Algorithm for the Scheduling of Virtual Network Functions

The scheduling of Virtual Network Functions (VN-Fs) is an important problem for Network Function Virtualization (NFV) resource allocation. In this paper, we investigate how to manage the Network Functions (NFs) efficiently to enhance the utilization of network resources. In the system model, we take into account the VNF transmission delay and processing delay at the same time. Our objective is to minimize the total end-to-end delay for all network services. To reduce the complexity of this issue, we propose a novel algorithm based on genetic algorithms by improving the method of crossover and mutation. The simulation results show that the proposed algorithm can reduce the total end-to-end delay at most 16.74%.