Load Balance Traffic Scheduling for Live Streaming Services in Edge
Hui Yan, Yaqian You, Gaoyu He, Zhengyi Zhong, Haoran Ji, Lasheng Kuang, Xiaomin Zhu · 2022
With the rapid development of smartphones and wireless communication networks, live streaming has gradually become a hot topic in recent years. For service providers, im-proving user live broadcast experience while reducing operating costs is the key to improving enterprise competitiveness. Load balancing is the key to achieving this goal. We model the real-time traffic scheduling problem as a bipartite graph matching problem, thereby automatically removing user geographic location constraints. Taking load balancing as the optimization goal, comprehensively considering the user's Qos constraints and the capability constraints of edge nodes, an optimization model is established. We propose a two-stage optimization method, namely LBS, which combines proportional allocation with heuristic optimization to improve problem solving efficiency. Through simulation experiments, the performance of LBS is effectively evaluated. Compared with the two baseline algorithms, LBS can significantly improve load balancing with a small time cost.