考虑截止资源公平调度的雾计算任务卸载  

Offloading of Fog Computing Tasks Considering Fair Scheduling of Cut-off Resources

作　　者：那勇 王明华 NA Yong;WANG Ming-hua(Jilin Distance Education Technology Innovation Center,Changchun 130022,China;Department of Distance Education Technology,The Open University of Jilin,Changchun 130022,China)

机构地区：[1]吉林省远程教育技术科技创新中心,吉林长春130022 [2]吉林开放大学远程教育技术中心,吉林长春130022

出　　处：《控制工程》2022年第12期2343-2348,共6页Control Engineering of China

基　　金：吉林省科技发展计划项目(20190902010TC);吉林省教育科学“十三五”规划重点项目(ZD19156)。

摘　　要：为解决雾计算网络中不同延迟期限的任务卸载问题,合理分配计算资源,提出一种考虑截止资源公平调度的雾计算任务感知卸载算法。首先,提出考虑计算资源公平的调度策略,确保在各自截止时间内完成的任务数量实现最大化,同时使网络具有很强的稳定性;然后,利用李雅普诺夫漂移加惩罚函数对任务队列长度进行调度,并设计调度策略决定要卸载到欠负载雾节点的任务数量,以充分利用网络中所有雾节点提供的计算资源;最后,仿真实验结果表明,该雾计算任务调度策略要优于选取的对比任务调度策略。In order to solve the problem of task offloading with different delay periods in fog computing networks and allocate computing resources reasonably,a task perception offloading algorithm considering fair scheduling of cut-off resources is proposed in this paper.Firstly,a scheduling strategy considering the fairness of computing resources is proposed to maximize the number of tasks completed within their respective deadline,and make the network have strong stability at the same time.Then,the length of the task queue is scheduled by using Lyapunov drift plus penalty function,and the scheduling strategy is designed to determine the number of tasks to be offloaded to the underloaded fog nodes,so as to make full use of the computing resources provided by all fog nodes in the network.Finally,the simulation results show that the proposed task scheduling strategy of fog computing is better than the selected comparative task scheduling strategies.

关 键 词：计算资源 公平调度 雾计算 任务卸载 感知决策 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]