<![CDATA[The rapid growth of cloud computing, large-scale data centers, and heterogeneous network traffic has exposed structural limitations in traditional distributed routing architectures. Conventional switching and routing mechanisms often lack global network visibility, resulting in suboptimal path selection, inefficient bandwidth utilization, and delayed convergence under dynamic traffic conditions. This study aims to design and evaluate a Software Defined Networking (SDN)-based optimization framework to enhance switching and routing performance through centralized programmability and adaptive traffic engineering. A quantitative experimental design was employed using network emulation across small-, medium-, and large-scale topologies. Comparative analysis was conducted between conventional routing protocols and the proposed SDN-based model. Performance metrics included throughput, end-to-end delay, packet loss rate, convergence time, and bandwidth utilization efficiency. Inferential statistical testing was applied to validate performance differences. Results demonstrate statistically significant improvements under the SDN framework, including increased throughput, reduced latency, lower packet loss, and faster failure recovery. Performance gains were more pronounced in large-scale and high-traffic scenarios, indicating strong scalability and resilience characteristics. The findings confirm that centralized control architecture fundamentally enhances routing optimization and network adaptability. SDN-based approaches provide a scalable and efficient solution for modern programmable network infrastructures.]]>
