摘要
云原生计算基于低开销容器化的运行方式非常契合边缘计算,因此,提出将云原生技术应用于边缘计算,发挥云原生的优势,使资源管控对应用开发部署透明化。考虑相较于云计算,边缘计算具有资源广分布、高异构、多碎片特征,亟须算网协同的资源管控。根据云原生相关技术的发展现状,整合软件定义网络与网络功能虚拟化等未来网络技术,提出全栈式云原生边缘计算架构。在此基础上,考虑容器的层次化特性,提出适用于有限边缘计算资源的低开销容器部署方法,分析探讨云原生边缘计算所面临的挑战。
Cloud native computing,featured by low-cost container technology,well fits edge computing.It was proposed to apply cloud native computing to make edge computing resource management and control transparent to application development and operation.Compared with cloud computing,edge computing resources are widely distributed,highly heterogeneous,and fragmented,which called for collaborative resource management and control.According to the development status of cloud native related technologies,through the integration of future networking technologies such as soft were defined network and network function virtualization,a full-stack cloud native based edge computing architecture was proposed.Then,considering the hierarchical characteristics of containers,a low-overhead container deployment optimization problem for resource-limited edge computing was studied.Finally,the development challenges faced by cloud native based edge computing were discussed.
作者
曾德泽
陈律昊
顾琳
李跃鹏
ZENG Deze;CHEN Lvhao;GU Lin;LI Yuepeng(School of Computer Science,China University of Geosciences,Wuhan 430074,China;School of Computer Science and Technology,Huazhong University of Science and Technology,Wuhan 430074,China)
出处
《物联网学报》
2021年第2期7-17,共11页
Chinese Journal on Internet of Things
基金
国家自然科学基金资助项目(No.61772480,No.61972171)
之江实验室开放课题(No.2021KE0AB02)。
关键词
云原生
边缘计算
容器
微服务
cloud native
edge computing
container
micro service
