UDC 004.72
NEURAL NETWORK MULTIPATH ROUTING IN SOFTWARE DEFINED NETWORKS BASED ON ANT COLONY OPTIMIZATION ALGORITHMS
D. A. Perepelkin, Dr. Sc. (Tech.), Full Professor, Dean of the Faculty of Computer Engineering, RSREU, Ryazan, Russia;
orcid.org/0000-0003-4775-5745, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
V. T. Nguyen, post-graduate student, RSREU, Ryazan, Russia;
orcid.org/0000-0003-2930-5775, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
Currently, the widespread demand for the implementation and use of various cloud solutions is a modern trend and a driving force for the development of network technologies. The growth of cloud application services provided through data centers with different network traffic needs demonstrates the limitations of traditional routing and load balancing methods. The integration of software defined networking (SDN) technology and artificial intelligence (AI) methods provides effective resource management and operation of computer networks. In this paper, an approach to neural network multipath routing in SDN based on ant colony optimization algorithms is proposed. An architecture and model of an artificial neural network are developed to solve the problem of multipath routing in SDN, which is capable of predicting the shortest paths based on communication link metrics. To optimize the hyperparameters of the neural network model, the paper proposes to use the algorithms of the ant system and the ant colony system. A visual software system SDNLoadBalancer has been developed and an experimental topology of the SDN has been designed, allowing a detailed study of the processes of neural network multipath routing in the SDN based on the proposed approach. The results obtained show that the proposed neural network model has the ability to predict routes with high accuracy in real time, which allows implementing various load balancing schemes in order to increase the performance of the SDN.
Key words: : software defined networks, neural network routing, multipath routing, deep learning, recurrent neural networks, swarm intelligence, ant colony optimization.