期刊名称：IAENG International Journal of Computer Science
印刷版ISSN：1819-656X
电子版ISSN：1819-9224
出版年度：2021
卷号：48
期号：1
语种：English
出版社：IAENG - International Association of Engineers
摘要：With the rapidly deployment of 5G communication, network functional virtualization has became one of key technology. However, how to achieve virtual network function service chaining (VNF-SC) adaptively and cost-effectively in an inter datacenter elastic optical network (interDC-EON) has become an interesting and challenging problem. In addition, network function virtualization has became one of the key technology for 5G Commons. In this paper, we propose a resource scheduling framework based on artificial intelligence. In addition, an effective VNF-SC deployment algorithm based on deep reinforcement learning (RL) is designed. In this algorithm, VNF-SC deployed scheme is obtained by using deep Q-learning according to the state of the network and the reward of the action. To verify the efficient of the proposed algorithm, a large number of experiments have been conducted. Experimental results demonstrate that the RL achieves better performance than several benchmarks, in terms of balancing the tradeoff among the overall resource utilization, the vNF-SC request blocking probability.
关键词：Virtual Network Function (VNF);VNF-service chain (VNF-SC);Reinforcement learning (RL);5G Communication
