Research On Adaptive Offloading Algorithm Of Mobile Edge Cloud Computing Based On Lyapunov Optimization

With the rapid development of the Internet of Things and 5G technology,intelligent mobile terminal devices incorporated into the Io T are increasing at an explosive rate and are playing an increasingly important role in every aspect of our daily life.Intelligent mobile terminal devices have limited capacity in terms of computing resources,storage capacity and battery energy,which cannot meet the high computing power,low response delay and high broadband required by intensive applications running on mobile terminal devices.This resource gap can be filled by remote centralized mobile cloud computing(CMCC),but it is not ideal in terms of low response latency and high power consumption costs.Driven by Io T and 5G communications,CMCC is shifting to MEC,which can address the shortcomings of the CMCC.In recent years,with the rapid development of resource-intensive applications,a lot of intensive applications such as virtual reality,augmented reality,multimedia delivery and artificial intelligence have emerged,which challenges the delay of traditional cloud computing and the computing power of edge computing.Higher requirements are put forward for response latency and data storage.In order to deal with the above challenges,Mobile Edge Clouds(MECs)are proposed.Mobile edge cloud computing can overcome the resource limitation of edge devices by providing high-performance cloud computing capability through wireless access network near the edge of mobile users.At present,profit and energy consumption cost from the perspective of edge system are urgent problems for MEC to solve.Therefore,this paper studies the profit of edge server of edge system and the optimization of energy consumption cost of edge cloud system from two perspectives.The main work of this paper is as follows:1.First of all,this paper introduces the evolution of edge computing and the overall structure of the MEC.Secondly,this paper introduces the basic idea of edge computing and the common computing offloading technology and the main application scenarios and common deployment schemes of MEC.Lastly,the basic theory of the queue involved in response delay and energy consumption cost is introduced,which provides a theoretical basis for the adaptive unloading of the back edge calculation.2.This paper studies the profit problem of the edge server in the middle layer under the framework of edge cloud system.Based on Lyapunov optimization,we propose an online algorithm with low complexity from the edge server point of view,using constraints to ensure that the statistical average response time of the offloading task does not exceed the maximum delay that the user can tolerate.Based on the stability of the system,the maximum delay that the user can tolerate,the fluctuating processing cost of the cloud server and the computing state of the edge server,the edge server can decide the processing location of the task(edge or cloud),so that the edge server can obtain the maximum profit.On the basis of the previous work,the energy consumption cost of the edge system during the task processing is studied under the framework of the three-layer edge cloud system.On the basis of Lyapunov optimization,we comprehensively consider the mixed power supply of the edge system by energy storage device,renewable energy and traditional power grid,the time-varying electricity price of traditional power grid,the tolerated time delay of mobile users,and the computing capacity of the edge system.On the premise that the response time of task processing does not exceed the user's tolerable delay,an appropriate decision should be made for the tasks unloaded by the mobile end user to be processed in the edge system or cloud data center,so as to minimize the cost of task processing in the edge system.