Research On Task Cache And Migration Strategy In Mobile Edge Computing Environment

The rapid development of mobile Internet industry has promoted the birth and development of Internet of Things,Big Data,Cloud Computing,Artificial Intelligence and other technologies.It is also the rapid development of these technologies that has brought profound changes to the Internet industry.All of these technologies have strict requirements on response time,equipment endurance and network environment,which is undoubtedly a huge challenge to the traditional network computing model.As a new computing paradigm,Mobile Edge Computing,(MEC)has been proposed as the most promising solution to the above needs.Its low latency,localization,close range and location awareness make the functions of various mobile applications greatly improved.Edge computing refers to a new computing mode for computing network edges.Its "edge" is a relative concept.Compared with the traditional "centralized computing",terminal devices with certain computing and caching capabilities in edge computing environment can participate in computing,which makes a large number of computing and caching resources of terminal devices effectively utilized under explosive growth.In addition,because mobile edge computing is closer to the source of data and closer to users,users are closer to the servers that provide computing in both network distance and space distance,thus meeting the needs of low latency.However,how to deploy and migrate services to make the best use of the advantages of mobile edge computing is also an urgent problem to be solved.Therefore,this paper studies mobile edge computing in two aspects: service caching and task migration.In this paper,the mobile edge computing system is composed of multiple mobile users and multiple servers.Its servers include edge servers,proxy servers and remote central servers.Mobile users send task requests to servers in the same system,and expect to get the result of task calculation in a relatively short time.Since mobile users in the same environment tend to request the same or similar computing tasks,if users are responded every time the repeated computing tasks,it will undoubtedly cause a great waste of computing and energy resources,and it is also a burden for the network situation in peak period.Therefore,researchers have derived task caching technology based on task migration technology.Cache some tasks to the edge server,and return the calculation result directly when the corresponding task request arrives,which not only reduces the response time of the task,but also reduces the energy cost of the user.This paper takes into account the influencing factors such as task utility and server access,and proposes a task caching strategy for the multi-server environment through long-term observations of past task requests and services in the system.The strategy establishes a mathematical model with the goal of minimizing the task response time of all users in the same edge system and proposes a heuristic algorithm to solve the problem.Finally,the task deployment decision is obtained,and the deployed environment is applied to short-term task migration.In technology.Finally,the experimental results are compared with the theoretically optimal calculation method.The simulation results show that the proposed heuristic algorithm effectively reduces the response time of user computing tasks and is close to the theoretical optimal computational strategy in performance.