Lstm-based server and route selection in distributed and heterogeneous SDN network

Abstract

Nowadays, Software-defined Networking (SDN) is increasingly being deployed in net-work systems and by network operators due to its benefits, such as automation leading to excellent reliability, more efficient network management, cost-savings, and faster scalability. The most common deployment architecture for SDN is a distributed system consisting of many independent domains, each controlled by an SDN controller. One of the well-known applications in SDN is server selection and routing. However, deploying server and route selection in distributed and heterogeneous SDN networks faces two issues. Firstly, the lack of global views of the entire system is due to the absence of standardized inter-communication between SDN domains for the distributed and heterogeneous SDN network. To address this issue, this paper utilizes our previous work, the open East-West interface SINA, to adaptively ensure network state consistency in a multi-domain SDN networks. Secondly, selecting the path for packet transmission based solely on the current network states of a local SDN domain is ineffective since the system cannot respond to unexpected changes in the network state. Predicting the link cost of the entire routing path is necessary. Therefore, this paper proposes an LSTM-based link cost prediction for the server and route selection mechanism in a distributed and heterogeneous SDN network. The experimental results demonstrate that our proposal improves link utilization, packet loss rate, response time, and overhead by up to 15%, 10%, 14%, and 25% respectively to benchmarks.