Research On Cache Placement And Task Scheduling Methods Based On Comprehensive Utility In Edge Environment

With the convergence of the Internet of Things,5G and artificial intelligence,diversified application scenarios and the connection of massive devices have higher requirements for the network,and users are increasingly demanding ultra-low latency and high-quality services.However,the computing power and storage resources of mobile devices are limited,which leads to the increasingly prominent contradiction between computing-intensive applications and resource-constrained mobile devices.Cloud computing can provide almost infinite resources for end users,but long-distance transmission will lead to higher service delay.Therefore,edge computing emerged as the times require.Edge computing is a new type of service that deploys servers close to end users or data sources,and can meet high concurrency and low latency.In the edge computing environment,a good cache placement strategy and a task scheduling strategy based on cache placement can reduce the network delay of users' requests,improve the resource utilization rate,and upgrade the service experience of users.Therefore,on the basis of existing research,the optimization method of cache placement and task scheduling based on comprehensive utility in edge computing environment is studied in this paper,in order to further improve the service performance of edge servers.The main research contents and innovations are as follows:(1)In view of the fact that the current caching strategy does not comprehensively analyze and consider the cache objects and server characteristics,resulting in low server space utilization and frequent data replacement,.In this paper,a cache placement optimization method based on comprehensive utility in edge computing environment is proposed.In this method,firstly,the popularity of data blocks,the remaining validity ratio of data blocks,the replacement rate of edge servers and other placement factors are quantified to establish the cache value model of data blocks.Then,according to the data block cache value,data block acquisition cost,data block placement cost and replacement cost,a comprehensive utility model of cache placement is obtained.Finally,the cache placement algorithm based on improved tabu search is used to solve the optimal placement of data blocks.The experimental results show that the proposed algorithm improves the performance of cache service rate,data response time and replacement number compared with DAH algorithm,GPA algorithm and NVCP algorithm.(2)Aiming at the problems that the existing task scheduling mechanism seldom considers cache resources and disk resources,resulting in low resource utilization and long task execution time,a task scheduling optimization method based on data cache is proposed in this paper.In this method,firstly,quantify the three factors of task and container correlation,task scheduling priority and data transmission cost,and design a task scheduling model based on data cache;Then combine this model with a weighted bipartite graph to establish a task scheduling optimization model based on weighted bipartite graph,in which the comprehensive values of the three objective factors are solved by the linear weighted sum of the edge weights;Finally,the Kuhn-Munkres algorithm based on weighted bipartite graph is used to achieve the optimal matching between tasks and containers,and the optimal task scheduling scheme is obtained.The experimental results show that the method improves the utilization of cached data,reduces the transmission of data,and improves the efficiency of task execution.