SDN Load Prediction Algorithm Based on Artificial Intelligence

5G/IMT-2020 networks have to provide new technical requirements for realizing new services such as Tactile Internet, medical services and others. 5G infrastructure will be based on Software-Defined networking and Network Function Virtualization for providing new quality level. In general, a significant number of the available Internet services and applications require exact value of network parameters such as latency, jitter, RTT and bandwidth. The SDN-based technologies should be able to control and manage dynamic QoS for different new services, which are a time constraint. For this reason, SDN-controller, like the main element of network infrastructure, must be stable and protected from external different threats. There are many works were on this task. Most of these works are goaled on stress tests of hardware and software parts, also one of the de-facto tests for each controllers is - generating OpenFlow “packetin” message from special traffic generator. Nevertheless, in “life mode” controller can be loaded differently, for example, uneven service load. We cannot build in advance various theoretical models of the controller load. In this regards, there is a need to develop a new approach for monitoring and prediction algorithm for build predicted models of OpenFlow activities. Also, this algorithm has to be independent of the hardware features of the controller and another technical integration peculiarities. In this paper proposed a novel approach for SDN load prediction based on artificial intelligence algorithms and totally monitoring of OpenFlow channels activities. Also in this paper, the possibility justification for predicting the load on hardware part, with the help of OpenFlow thread analytics was given. © 2019, Springer Nature Switzerland AG.

Volkov A.1 , Proshutinskiy K.1 , Adam A.B.M.2 , Ateya A.A. 1, 3 , Muthanna A. 1, 4 , Koucheryavy A. 1
Springer Verlag
1141 CCIS
  • 1 The Bonch-Bruevich State University of Telecommunications, Saint Petersburg, Russian Federation
  • 2 School of Communications and Information Engineering, Chongqing University of Posts and Telecommunications, Chongqing, China
  • 3 Electronics and Communications Engineering, Zagazig University, Zagazig, Ash Sharqia Governorate 44519, Egypt
  • 4 Peoples’ Friendship University of Russia (RUDN University), 6 Miklukho-Maklaya Street, Moscow, 117198, Russian Federation
Ключевые слова
5G; Artificial intelligence; Prediction; SDN
Дата создания
Дата изменения
Постоянная ссылка

Другие записи