子任务调度和时延联合优化的MEC卸载方案  

MEC offloading scheme based on joint optimization of subtask scheduling and delay

作　　者：陈韩 张晶[1,2,3] 董俊 董洁[6] CHEN Han;ZHANG Jing;DONG Jun;DONG Jie(College of Telecommunication&Information Technology,Nanjing University of Posts and Telecommunications,Nanjing 210003,China;Jiangsu Key Laboratory of Wireless Communications,Nanjing University of Posts and Telecommunications,Nanjing 210003,China;Institute of Internet of Things,Nanjing University of Posts and Telecommunications,Nanjing 210003,China;Institute of Intelligent Machines of Hefei Institute of Physical Science,Chinese Academy of Sciences,Hefei 230031,China;Anhui Zhongke Deji Intelligent Technology Co.Ltd.,Hefei 230031,China;The State Radio Monitoring Center,Beijing 100037,China)

机构地区：[1]南京邮电大学通信与信息工程学院,江苏南京210003 [2]南京邮电大学江苏省无线通信重点实验室,江苏南京210003 [3]南京邮电大学物联网研究院,江苏南京210003 [4]中国科学院合肥物质科学研究院智能机械研究所,安徽合肥230031 [5]安徽中科德技智能科技有限公司,安徽合肥230031 [6]国家无线电监测中心,北京100037

出　　处：《系统工程与电子技术》2023年第2期572-579,共8页Systems Engineering and Electronics

基　　金：国家重点研发计划(2020YFB1807202);国家自然科学基金(92067201);江苏省重点研发计划(BE2020084-1);江苏省自然科学基金(BK20130875);南京邮电大学校级科研基金(NY219044)资助课题。

摘　　要：移动边缘计算(mobile edge computing,MEC)为5G超低时延业务提供了解决方案。如何设计低时延、高效率的任务卸载方案,是MEC面临的主要难题之一。为此,针对端边协同MEC服务场景,研究了大型计算任务的低时延、低能耗部分卸载方案,通过将用户任务划分为多个有顺序依赖关系的子任务并构建子任务的有向无环关系图,设计了能够最小化卸载时延的子任务调度方案,提出了基于任务复制的最早卸载执行算法,解决了能耗受限下的任务最小时延卸载计算。仿真结果表明,提出的MEC卸载方案能够有效减少任务处理时延,降低系统能耗。Mobile edge computing(MEC)provides solutions for 5G ultra-low latency services.How to design a low-latency and high-efficiency task offloading program is one of the main problems faced by MEC.To this end,the low-latency,low-energy partial offloading scheme of large-scale computing tasks is studied for the end-side collaborative MEC service scenario,and the user task is divided into multiple subtasks with sequential dependencies and directed acyclic graph for subtasks is constructed.A subtask scheduling scheme that can minimize the offload delay is designed,and the earliest offload execution algorithm based on task replication is proposed,which solves the task of minimum delay offload calculation under energy constraints.The simulation results show that the proposed MEC offloading scheme can effectively reduce the task processing delay and reduce system energy consumption.

关 键 词：移动边缘计算 部分卸载 有向无环图 子任务调度 时延最小化 

分 类 号：TN929.5[电子电信—通信与信息系统]