A Dynamic Load Balancing Strategy for Routing Selection in Multi-Service Networks

Quality of Service (QoS) routing aims to maximize the use of network resources while subject to a series of QoS constraints. Due to the progressively intricate network environments and escalating demands for bandwidth, how to provide efficient multi-service scheduling and routing is crucial for improving network QoS. However, it is challenging to take both routing and network bandwidth resources into account in a global view. Motivated by these challenges, a method under the mixed integer linear programming (MILP) model is proposed firstly to find the optimal solution for routing selection, which is efficient in multi-service networks. Consequently, a heuristic algorithm based on load balance is proposed to make the scheduling problem tractable while keeping the solution quality satisfactory. The main idea of the heuristic algorithm is to achieve load balance in multi-service networks by dynamically sorting multiple services and selecting appropriate routes from multi-alternative paths. Extensive simulations demonstrate that satisfactory results can be obtained with low computational complexity for both small-scale and large-scale networks.