Cycle-based local update method for edge-load balancing problem

In the management network with SDN controllers, a traffic load-leveling method for communication links was considered to avoid congestion on the links for network service. Our study has modeled the network as a flow network, defined the ratio of flow to capacity in an edge as a load factor, and formulated the network flow optimization problem for finding the flow path to minimize the maximal load factor in the entire network as the unsplittable flow edge-load factor balancing problem. Previous work proposed the cycle-based heuristic approach for this problem while this problem is NP-hard. However, the proposed method has a time consumption issue about all combinations of flow routing. Therefore, this paper proposes efficient local update methods in the cycle-based heuristic approach by using meta-heuristic strategies.