DRSIR: A Deep Reinforcement Learning Approach for Routing in Software-Defined Networking
preprintposted on 01.05.2021, 10:23 by Daniela Casas Velasco, Oscar Mauricio Caicedo Rendon, Nelson Luis Saldanha da Fonseca
Traditional routing protocols employ limited information to make routing decisions which leads to slow adaptation to trafﬁc variability and restricted support to the quality of service requirements of the applications. To address these shortcomings, in previous work, we proposed RSIR, a routing solution based on Reinforcement Learning (RL) in SoftwareDeﬁned Networking (SDN). However, RL-based solutions usually suffer an increase in the learning process when dealing with large action and state spaces. This paper introduces a different routing approach called Deep Reinforcement Learning and SoftwareDeﬁned Networking Intelligent Routing (DRSIR). DRSIR deﬁnes a routing algorithm based on Deep RL (DRL) in SDN that overcomes the limitations of RL-based solutions. DRSIR considers path-state metrics to produce proactive, efﬁcient, and intelligent routing that adapts to dynamic trafﬁc changes. DRSIR was evaluated by emulation using real and synthetic trafﬁc matrices. The results show that this solution outperforms the routing algorithms based on the Dijkstra’s algorithm and RSIR, in relation to stretching (stretch), packet loss, and delay. Moreover, the results obtained demonstrate that DRSIR provides a practical and viable solution for routing in SDN.