Journal of Frontiers of Computer Science and Technology ›› 2023, Vol. 17 ›› Issue (10): 2325-2342.DOI: 10.3778/j.issn.1673-9418.2303042

• Frontiers·Surveys • Previous Articles     Next Articles

Survey of Task Offloading Technology in Cloud-Edge Resource Collaboration

TIAN Yumeng, LIU Zhibo, ZHANG Kai, LI Zhongbo, XIE Yongqiang   

  1. 1. Institute of Systems Engineering, Academy of Military Sciences, Beijing 100141, China
    2. Equipment Project Management Center of the Equipment Development Department of the Military Commission, Beijing 100009, China
  • Online:2023-10-01 Published:2023-10-01

云边资源协同中的任务卸载技术综述

田雨萌,刘志波,张凯,李忠博,谢永强   

  1. 1. 军事科学院 系统工程研究院,北京 100141
    2. 军委装备发展部 装备项目管理中心,北京 100009

Abstract: With the development of wireless communication technology and the Internet of things, the Internet of everything is becoming a reality and the number of connected devices is growing exponentially. Traditional cloud computing has powerful computation, storage, and network resources, but uncontrollable service delays will occur in the face of surges in data traffic. Edge computing places resources closer to the terminal, but the storage capacity of edge devices is small, and the processors equipped with them usually have weak computing power, so edge computing has low latency but limited resources. By combining the advantages of both cloud computing and edge computing, cloud-edge collaboration can effectively improve resource service capacity and service quality. It has a broad development prospect. Resource collaboration is an important aspect of cloud-edge collaborative service capabilities. Task offloading technology is one of the key technologies of cloud-edge resource collaboration. To promote the future development of this field and inspire researchers, the task offloading technology in cloud-edge resource collaboration is analyzed. Firstly, this paper sorts out the development history of cloud-edge collaboration and introduces the concept connotation of cloud-edge resource collaboration and task offloading, as well as the application scenarios of cloud-edge resource collaboration. Then, this paper summarizes the development of this technology at home and abroad from three aspects: uninstall objects, uninstall granularity, and service quality evaluation indicators. Finally, this paper proposes future development directions of task offloading technology in cloud-edge resource collaboration.

Key words: cloud-edge collaboration, resource collaboration, cloud computing, edge computing

摘要: 随着无线通信技术和物联网技术的发展,万物互联正在成为现实。传统云计算具有强大的计算、存储、网络等资源,但面对激增的数据流量仍会产生不可控的服务延迟。边缘计算将资源放置在更靠近终端的位置,但边缘设备的存储容量较小,配备的处理器通常算力较弱,因此边缘计算延迟低却资源受限。云边协同通过综合云计算和边缘计算两种计算范式的优势,可以有效提升资源服务能力和服务质量,具备广阔的发展前景。资源协同是云边协同服务能力的一个重要方面。任务卸载技术是云边资源协同的关键技术之一,为了进一步促进这一领域今后的发展,启发研究人员的思路,对云边资源协同中的任务卸载技术进行梳理分析,首先,梳理了云边协同的发展历程并介绍了云边资源协同和任务卸载的概念内涵及云边资源协同的应用场景;然后,从卸载对象、卸载粒度、服务质量评价指标三方面对国内外该技术的发展进行了归纳整理;最后,总结并提出了云边资源协同中的任务卸载技术未来发展的方向。

关键词: 云边协同, 资源协同, 云计算, 边缘计算