Maximizing Network Utilization for SDN Based on WiseAnt Colony Optimization

Software-Defined Network (SDN) allows a centralized controller to globally plan packets forwarding according to the operator's objectives. It may require more forwarding rules, and multi-path routing which is allowed in SDN to improve network utilization and throughput means more forwarding rules. However, the forwarding table in TCAM-based SDN switch is an important but limited resource. With this limitation, to satisfy global network objective, such as maximum flow, we formulate this problem as the Bounded Forwarding-Rules Maximum Flow (BFR-MF) problem which is NP-hard. We maintain fairness among flows to guarantee the performance of individual flows and a certain level of QoS. We improve the Ant Colony Optimization algorithm, such as adding heuristic information based on forwarding table size, and propose the WiseAnt Colony Optimization (WACO) algorithm to solve the BFRMF problem. We use normalized maximum flow to evaluate the performance of WACO algorithm. Extensive simulations show that WACO algorithm performs better in network utilization on the premise of a certain level of QoS, especially for the data center network.