Optimizing Data Traffic with SDN for Link Failure Resilience Using Multi-Commodity Flow

Traditional routing algorithms, including those used in legacy IP-based networks and early Software-Defined networking (SDN) implementations, often underestimate the impact of link failures. These systems lack the adaptability required for optimal routing during failure recovery, resulting in reduced network performance and reliability. Although SDN-based traffic engineering has brought significant advancements in load balancing and throughput optimization, it frequently fails to address the challenges posed by link failures. In this paper, we address this gap by incorporating link failure probabilities into a multicommodity flow model, designed to optimize routing decisions dynamically. By integrating failure risks directly into the traffic engineering process, our solution significantly improves the network's resilience and reliability, even under adverse conditions. This method enhances network performance in dynamic SDN environments and provides a more robust foundation for maintaining seamless data flow during unexpected disruptions, ensuring more efficient and effective traffic optimization.