Research And Implementation Of Computation Offloading And Image Caching In Edge Computing Platform

With the rapid development of information technology,the increasing amount of data has brought challenges to the traditional cloud computing model.To this end,the edge computing model that uses network edge nodes to process and analyze data emerges as the times require,and complements the traditional cloud computing model.However,edge devices tend to be lightweight,and how to make rational use of limited computing resources and storage resources becomes an important problem that edge computing needs to solve.Aiming at addressing the above problems,this paper adopts the idea of computing offloading and image caching,and proposes a computing offloading method based on bilateral game and an image caching method based on popularity prediction.Specifically,the offloading method adopts the method of reaggregation of associated tasks to obtain a new task group,and then the task group and the cloud are matched by a many-to-one bilateral game,thereby realizing the computing offloading.For the caching method,the GRU-based image popularity prediction strategy is adopted to obtain the prediction result of the popularity,and then the result is used as a basis for the cache replacement.Furthermore,by combining with the Kubernetes Federation platform,the scheduler and cache of the edge computing cloud platform are designed,thus implementing the cloud-edge integrated edge computing cloud platform.Finally,the effectiveness of the scheduler and cache is verified by comparison experiments.