Route Optimization in SDNs: Incorporating Link Quality Estimation and Critical Switch Awareness

Abstract

Software-Defined Networks (SDNs) have revolutionized network architecture, centralizing control, and introducing unparalleled levels of flexibility and programmability. Within this transformative landscape, the optimization of routing has emerged as a fundamental element in network management, a crucial factor in fully realizing the potential of SDNs. This study introduces an innovative approach that seamlessly integrates two critical components—link quality estimation and critical switch awareness—into the route path selection process, with the primary objective of elevating forwarding performance within SDNs. A noteworthy development in route selection is the concept of routing based on the identification of critical switches. At the heart of this research lies the imperative question: How can routes be optimized effectively and dynamically within the dynamic realm of SDNs? Implementing link quality estimation stands out as a pivotal catalyst for the enhancement of network performance. Our method consistently elevates the efficiency of data transmission, reduces latency, and upholds elevated levels of Quality of Service (QoS) by selecting routes based on real-time assessments of crucial link metrics, including latency, packet loss, and available bandwidth. This work introduces a multi-constraint iteration of Route Path Selection Optimization (RPSO), which integrates parameters such as Link Latency and Link Delivery Ratio, both in relation to the links and the switches themselves. These parameters advocate for a route selection strategy that prioritizes routes with a reduced count of critical switches and an emphasis on high link quality. The experimental results are compelling, showcasing a substantial 55.73% increase in throughput when utilizing the proposed scheme. Furthermore, the packet delivery ratio experiences a notable increase of 12.5%, while the path stretch is appreciably reduced by 37%.