Computer Integrated Manufacturing System

Previous Articles     Next Articles

Edge computing resource scheduling overview: Historical perspective,architecture,modeling,and method analysis

ZHOU Xu1,MIAO Hui2,YANG Jing1+,JIANG Wu2,LIAO Xiaoyan2,LI Yijun2,LI Shaobo1,LU Jialin1   

  1. 1.State Key Laboratory of Public Big Data,Guizhou University
    2.BAISAHNCLOUD

边缘计算资源调度:历史、架构、建模与方法分析

周绪1,苗辉2,杨静1+,江武2,廖晓燕2,李逸骏2,李少波1,鲁加林1   

  1. 1.贵州大学省部共建公共大数据国家重点实验室
    2.贵州白山云科技有限公司

Abstract: Edge computing task offloading and resource scheduling is one of the current research directions that have attracted much attention in the field of computer science and information technology,and it is a promising research direction.At present,no literature has been published to comprehensively compare and summarize the performance of edge computing task offloading and resource scheduling methods in terms of delay,energy consumption,load,and scheduling efficiency,and to give the research progress,challenges,and development directions.To address the above issues,firstly,the development of edge computing technology is reviewed;secondly,the edge computing scheduling framework is summarized,and delay,energy consumption,resource load balancing and scheduling efficiency are computationally modeled as the indicators for evaluating the performance of algorithms,and the problem model is constructed based on the optimization of the corresponding indexes;then,edge computing scheduling methods are classified into classical and advanced methods for comparative analysis;then,the investigation of vehicle Networking and Autonomous Driving,Virtual Reality and Cloud Gaming,Smart City and Intelligent Transportation,Industrial Internet of Things and Intelligent Manufacturing,and Smart Life and Medical Edge;finally,the challenges and prospects are given in order to provide theoretical guidance and reference for researchers.

Key words: edge computing, resource scheduling, task offloading, performance modeling, scheduling methods

摘要: 边缘计算任务卸载和资源调度是当前计算机科学与信息技术领域中备受关注的研究方向之一,是具有广阔前景的研究方向.目前,尚未有文献对边缘计算任务卸载和资源调度的方法在时延、能耗、负载、调度效率方面的性能进行全面的对比与归纳,并给出研究进展、挑战及发展方向.针对以上问题,首先,回顾了边缘计算技术的发展脉络;其次,总结了边缘计算调度框架,并将时延、能耗、资源负载均衡和调度效率作为评价算法性能的指标进行计算建模,并根据相应指标的优化构建问题模型;然后,将边缘计算调度方法分为经典方法和进阶方法进行对比分析;接着,调研车联网与自动驾驶、虚拟现实与云游戏、智慧城市与智能交通、工业物联网与智能制造、智慧生活与医疗边缘的应用研究现状;最后,给出目前存在的挑战和展望,以期为研究者提供理论指导和借鉴。

关键词: 边缘计算, 资源调度, 任务卸载, 性能建模, 调度方法

CLC Number: