Research On Cooperative Task Scheduling In Mobile Edge Computing

Mobile edge computing can provide users with high-bandwidth,low-latency services at the edge of the network by sinking the computing and storage capabilities of the cloud data center to the edge of the network,meeting the urgent needs of everincreasing mobile devices and increasingly dense computing tasks.This paper studies the problem of collaborative task scheduling based on user mobility scenarios and task dependencies in mobile edge computing,and constructs a two-tier collaborative task scheduling system model between mobile devices and edge servers,and between edge servers and edge servers.The model includes physical models such as mobile devices and edge servers,wireless communication models between mobile devices and base stations,local computing and edge computing models,and time and energy consumption models during task processing.Aiming at the problem of collaborative task scheduling based on user mobility scenarios,this paper takes as the optimization goal of minimizing the average execution time of tasks under the premise of ensuring system service quality,and proposes a heuristic optimization algorithm(MAHO algorithm)for user mobility perception.The MAHO algorithm builds a user mobility model based on the known user movement trajectory,and makes appropriate adjustments to the wireless communication model and the time and energy consumption model in the task processing process.In terms of scheduling strategy,the MAHO algorithm preferentially selects tasks with higher urgency for scheduling,according to the resource matching degree of the edge server,and combining two special scenarios(local computing scenario and access server computing scenario)to determine the task in the local Computing or edge computing and which edge server is used for edge computing.Aiming at the problem of collaborative task scheduling based on task dependency,this paper also takes as the optimization goal of minimizing the average execution time of tasks under the premise of ensuring system service quality,and proposes a taskdependent-aware heuristic optimization algorithm(DAHO algorithm).The DAHO algorithm is based on the MAHO algorithm and adds a task dependency model to ensure the dependency limit during task execution.In terms of scheduling strategy,the DAHO algorithm preferentially selects tasks with a smaller latest scheduling time and conforms to the dependency limit for scheduling.Like the MAHO algorithm,according to the resource matching degree of the edge server,two special scenarios are combined to determine the task scheduling.decision making.In the experimental link,this paper verifies the algorithm performance of the proposed MAHO algorithm and DAHO algorithm through simulation experiments and real platform experiments.Both the simulation experiment results and the real experimental results show that the MAHO algorithm and DAHO algorithm proposed in this paper are compared with the existing TAS algorithm and MAMTS algorithm have obvious advantages in the average execution time and overtime rate of tasks.