数据与计算发展前沿 ›› 2020, Vol. 2 ›› Issue (4): 80-91.doi: 10.11871/jfdc.issn.2096-742X.2020.04.007

• 专刊:下一代互联网络技术与应用(下) • 上一篇    下一篇

面向算力网络的边缘资源调度解决方案研究

李铭轩1,*(),曹畅1(),唐雄燕1(),何涛1(),李建飞1(),刘秋妍2()   

  1. 1.中国联通网络技术研究院,未来网络研究部,北京 100048
    2.中国联通网络技术研究院,无线技术研究部,北京 100048
  • 收稿日期:2020-03-28 出版日期:2020-08-20 发布日期:2020-09-10
  • 通讯作者: 李铭轩
  • 作者简介:李铭轩,中国联通网络技术研究院高级工程师,硕士,美国IEEE高级会员,中国电子学会高级会员,从事技术研发和标准跟踪工作。主要研究方向为大数据技术,云计算技术,业务平台技术和IT支撑系统技术;参与CCSA、GSMA、ITU等国内外标准组织的会议。
    本文中负责整体技术解决方案设计、技术架构设计。
    Li Mingxuan, master’s degree, senior engineer of China Unicom Network Technology Research Institute, senior member of the United States IEEE, senior member of the Chinese Institute of Electronics, engaged in technology research and development and standard tracking; the main research directions are big data technology, cloud computing technology, business platform technology And IT support system technology; participate in CCSA, GSMA, ITU and other domestic and international standards organizations meetings.
    In this paper, he is responsible for overall technical solution design, technical architecture design.
    E-mail: limx59@chinaunicom.cn|曹畅,中国联合网络通信有限公司网络技术研究院未来网络研究部,高级专家、智能云网技术研究室主任,博士后,主要研究方向为IP网宽带通信、SDN/NFV、新一代网络编排技术等。
    本文中完成了论文的算力网络技术架构和算力网络应用需求分析。
    Cao Chang, postdoctoral, senior expert of Future Network Research Department, Director of Intelligent Cloud Network Technology Research Department, Network Technology Research Institute, China United Network Communications Co., Ltd. His main research directions are IP network broadband communication, SDN / NFV, and next-generation network orchestration. Technology, etc.
    In this paper, he is responsible for the analysis of the overall technical architecture and computing power network application requirements.
    E-mail: caoc15@chinaunicom.cn|唐雄燕,中国联合网络通信有限公司网络技术研究院,教授,首席科学家、智能网络中心总架构师,北京邮电大学兼职教授、博士生导师,主要研究方向为宽带通信、光纤传输、互联网/物联网、SDN/NFV与下一代网络等。
    本文中负责设计算力网络整体架构和下一代网络边缘技术架构。
    Tang Xiongyan, Professor, is the Chief Scientist of China United Network Communications Co., Ltd., Institute of Network Technology, Chief Architect of Intelligent Network Center, Adjunct Professor and Doctoral Supervisor of Beijing University of Posts and Telecommunications. His research interests include broadband communications, fiber optic transmission, internet networking, SDN, NFV and next-generation networks.
    In this paper, he is responsible for designing the overall architecture of the computing power network and the next-generation network edge technical architecture.
    E-mail: tangxy@chinaunicom.cn|何涛,中国联合网络通信有限公司网络技术研究院,硕士,高级工程师,主要从事云化网络及数据通信网络相关技术研究。
    本文中负责云原生平台Kubernetes研究。
    He Tao, master and senior engineer of Network Technology Research Institute of China United Network Communications Co., Ltd., is mainly engaged in researches on cloud-based networks and data communication network related technologies.
    In this paper, he is responsible for dong researches on cloud native platform Kubernetes.
    E-mail: het21@chinaunicom.cn|李建飞,中国联合网络通信有限公司网络技术研究院,硕士,高级工程师,主要从事下一代网络、AI算法以及自动驾驶等相关技术研究。
    本文中负责边缘计算嵌入式架构、GPU架构研究和人工智能应用分析。
    Li Jianfei, master and senior engineer of Network Technology Research Institute of China United Network Communications Co., Ltd., is mainly engaged in researches on next-generation networks, AI algorithms, and autonomous driving.
    In this paper, she is responsible for researching edge computing embedded architecture, GPU architecture and artificial intelligence application analysis.
    E-mail: lijf299@chinaunicom.cn|刘秋妍,中国联通网络技术研究院,博士后,高级工程师,主要研究方向为无线通信与区块链技术。
    本文中负责研究容器标签资源调度机制。
    Liu Qiuyan, Ph.D., is a senior engineer of China Unicom Network Technology Research Institute. Her main research directions are wireless communication and blockchain technologies.
    In this paper, she is responsible for researching container tag resource scheduling mechanism.
    E-mail: liuqy95@chinaunicom.cn
  • 基金资助:
    国家重点研发计划(2019YFB1802800,2019YFB1802600)

Research on Edge Resource Scheduling Solutions for Computing Power Network

Li Mingxuan1,*(),Cao Chang1(),Tang Xiongyan1(),He Tao1(),Li Jianfei1(),Liu Qiuyan2()   

  1. 1. Department of Future Network Research, China Unicom Network Research Institute, Beijing 100048, China
    2. Department of Wireless Technology Research, China Unicom Network Research Institute, Beijing 100048, China
  • Received:2020-03-28 Online:2020-08-20 Published:2020-09-10
  • Contact: Li Mingxuan

摘要:

【目的】介绍了算力网络面向业务需求,在云、网、边之间按需分配和灵活调度计算资源的网络组织方式。该方式旨在实现更广泛的计算资源的统一调度和纳管。而在网络边缘侧,由于各种嵌入式数量众多,架构各异,现有的资源调度方式难以满足算力纳管的需求。【方法】从算力网络架构出发,基于云原生的资源调度机制,阐述了轻量化、多集群的分级边缘资源调度方案。【结果】基于轻量级的云原生平台,实现了面向算力网络的前端海量边缘设备的统一纳管,并且能够在多种架构的嵌入式平台进行部署。【局限】作为面向算力网络整体技术架构,“云、边、端”中前端设备的统一资源调度平台,如何实现云边协同、人工智能算法等在前端嵌入式集群中的实现和部署,使其前端设备更具自主性是其下一步需要解决的重要问题。【结论】面向算力网络的前端嵌入式资源调度方案可以广泛应用于物联网、车联网、智慧城市等领域,进一步实现前端设备的自主处理能力,解决我国智能产业领域创新能力和支撑不足等现实问题。

关键词: 算力网络, 容器技术, 嵌入式技术, 边缘计算

Abstract:

[Objective] This paper introduces a network organization method for computing power network to satisfy business needs, which is able to flexibly schedule and allocate computing resources among clouds, networks, and edge devices. This approach aims to schedule and manage a wider range of computing resources in a unified framework. At the edge of the network, due to the large number of embedded devices and their different architectures, it is difficult for the existing resource scheduling methods to meet the demand of computing power. [Methods] Starting from the computing network architecture and based on the cloud-native resource scheduling mechanism, a lightweight, multi-cluster hierarchical edge resource scheduling scheme is described. [Results] Based on the lightweight cloud-native platform, we successfully manage and deploy a massive amount of hetero-architecture edge devices inside computing power networks in a unified framework. [Limitations] As a unified resource scheduling platform for front-end equipment in the “cloud, edge, and end” designed for computing power networks, it is important to solve the problems of implementing cloud-side collaboration, deploying artificial intelligence algorithms in front-end embedded clusters and making front-end equipment more autonomy. [Conclusions] The front-end embedded resource scheduling solution for computing power networks can be widely used in the Internet of Things, Internet of Vehicles, smart cities and other fields to improve the autonomous processing capabilities of the front-end equipment, and solve other practical problems such as the lack of innovation capabilities and insufficient supports in intelligent industry of China.

Key words: computing power network, container technology, embedded technology, edge computing