Evaluation of framework for reliability model in software defined networking using hybrid meta-heuristic algorithm

Evaluation of framework for reliability model in software defined networking using hybrid meta-heuristic algorithm

연구 분야: Networking

학회: International Journal of Information Technology

Meta-heuristic algorithms have emerged as powerful tools in solving complex optimization problems integrated in software-defined networks (SDN). SDNs offer effective network management based on the separation of control and data planes, yet the problem arises in assuring reliability under dynamic traffic conditions. To address this issue, there exist two reliability frameworks that embed different hybrid meta-heuristic algorithms, the Taylor Aquila and the Aquila Wild Geese algorithms. The framework intends to improve traffic classification and ensure cluster-based multipath routing with fault tolerance and adaptation to dynamic environments. It compares both the frameworks in terms of delay and throughput for 100 nodes. The simulation showed that both algorithms improved network reliability. However, despite having similar working phases, it tends to favor the Aquila Wild Geese-based framework over the Taylor Aquila-based framework in terms of delay and throughput. The compared analyses show that the superiority of hybrid approaches has better performance and scalability for SDN. This work thus points to the capability of hybrid meta-heuristics in tackling the dynamic challenges posed by SDNs. Still, it points to further research in terms of real-world implementation and scalability.