The paper focused on cache optimization and service selection algorithms in the cloud-edge environment. In order to solve the problem of cached content in edge servers, factors such as energy consumption and cost are considered, and finally a cache optimization model based on popularity was proposed. As for service selection, a Q-learning-based service selection algorithm is proposed to address the problems of dynamic task allocation and the optimization of Qos in the edge computing environment. The experimental results show that the proposed cache optimization and service selection algorithms in cloud-edge environment can better improve the cache hit ratio, minimize the transmission overhead, and ensure the server load balancing in the cloud-edge environment.

本文重点研究云边缘环境中的缓存优化和服务选择算法。为解决边缘服务器缓存内容的问题，综合考虑能耗和成本等因素，最后提出基于流行度的缓存优化模型。在服务选择方面，针对边缘计算环境下动态任务分配和Qos优化问题，提出了一种基于Q-learning的服务选择算法。实验结果表明，提出的云边环境缓存优化和服务选择算法能够更好地提高缓存命中率，最大限度地减少传输开销，保证云边环境下的服务器负载均衡。