Our research group has been actively exploring resource scheduling in edge-cloud environments, aiming to improve service efficiency, scalability, and responsiveness in highly distributed and heterogeneous systems. Our work spans intelligent task allocation, context-aware scheduling, and adaptive autoscaling, with a strong focus on learning-based and collaborative optimization methods. (1) For practical contributions, we have implemented and deployed various scheduling strategies in containerized and Kubernetes-oriented platforms, addressing challenges such as SLA guarantees, burstable billing, and multi-tier orchestration in real-world, dynamic scenarios. (2) For theoretical contributions, we have applied mathematical theories, including the hypergraph concept and game theory, to mathematically model and address the fragmentation of resource-task supply and demand in scheduling optimization, ensuring a more efficient distribution of resources across the system. These research efforts have been supported by key national and industrial projects, including collaborations with the China Academy of Aviation Manufacturing Technology, Paiou Cloud Computing, and the State Grid Corporation of China. Through these collaborations, our technologies have been successfully transferred and applied in industrial settings such as smart manufacturing, edge-assisted energy systems, and edge platform design. Backed by funding from the National Natural Science Foundation of China and the National Key R&D Program, we have validated our work with large-scale real-world datasets and prototype deployments across diverse application domains. Our research has not only yielded a series of high-impact publications but also contributed practical solutions to pressing problems in edge intelligence and resource management, promoting the real-world adoption of intelligent scheduling technologies.
