SoftFlow: Distributed Control Plane Load Balancing in Softwarized 5G WANs using Flow Migration
Deborsi Basu, Vikash Kumar Gupta, Uttam Ghosh, Raja Datta · 2022 IEEE 13th Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON) · 2022
The next-generation wireless communication networks (e.g., 5G, 6G and beyond) are going to encounter excessive traffic overheads due to the rapid digitization of technologies. Most importantly, the context of the new normal is continuously enforcing all possible media and applications to shift into the online mode. Rapid digitisation and data aggregation cause excessive traffic overload inside a network. All the existing state-of-the-art technologies are inefficient to control such a sudden overshoot of network traffic. So, an effective network model is urgently required. Keeping this fundamental area in mind, in this work, we have implemented an adaptive load balancing model over distributed wide area network (WAN) using a multi-tier SoftFlow method that can dynamically dissipate the controller load (C-plane load) among the different hierarchical network clusters. We have proposed a dynamic OpenFlow switch migration technique that can work as a back-end load balancer without disrupting network traffic flow. The flexible flow management algorithm (SoftFlow) along with the vSDN-driven (virtualized Software-Defined Network) results in a satisfactory reduction in traffic load (20% on an average traffic flow). Managing a high traffic flow over a wide geographical region is primarily targeted but rescaling is possible with systematic and parametric optimizations. The model has been realized over a real network topology (India - ERNET). The comparative inspection with existing state-of-the-art unbalanced models is performed to validate the utility of the approach.