Research On Task Dispatching Algorithm For Mobile Edge Computing

In recent years,with the vigorous development of the mobile Internet,the growing popularity of mobile devices,such as smartphones and laptops,has led to a variety of different mobile applications.These applications have stringent requirements for service response time.But it's difficult to meet the high requirements of service quality for lowperformance mobile devices due to the limitation of the size.Mobile Edge Computing(MEC),is expected to be a promising solution to solve this dilemma.Efficient task offloading and resource allocation have been two challenging problems in MEC systemFor the task offloading problem of independent tasks in MEC system,in this paper,we propose an innovative scheme,that jointly solves the task offloading and resource allocation problem,optimize the overall cost of task execution.It's difficult to solve directly due to the high coupling between resource allocation and task offloading strategy in the modeled problem.To reduce the computation complexity,we decouples the optimization problem into two sub-problems:one of them solves the resource allocation problem while the other tackles the task offloading issue.The Lagrangian multiplier method is used to solve the resource allocation problem under a specific task offloading strategy.Based on this,we designed a task oflloading algorithm for joint resource allocation,minimize the total cost of task processing.Compared with related algorithms,our algorithm can reduce the total cost by up to about 27%For the task offloading problem of the tasks with dependencies in MEC,On the basis of comprehensive consideration of delay requirements and dependencies,we propose an algorithm,that solves the task offloading problem for the tasks with dependencies in MEC.Specifically,we calculate the priority of the dependent subtasks on the user device,then decide the offloading strategy according to the delay requirements of different user minimize the task execution time.Compared with related algorithms,our algorithm can reduce the processing time by up to about 8%.