数据与计算发展前沿 ›› 2022, Vol. 4 ›› Issue (6): 55-66.

CSTR: 32002.14.jfdc.CN10-1649/TP.2022.06.006

doi: 10.11871/jfdc.issn.2096-742X.2022.06.006

• 专刊：东数西算：开启算力经济时代的世纪工程（下） • 上一篇    下一篇

算力网络一体化支撑方案及应用场景探索

叶沁丹1,*(),范贵生2,黄衡阳1   

  1. 1.浩鲸云计算科技股份有限公司，商业发展部，广东 广州 511400
    2.浩鲸云计算科技股份有限公司，OSS产品线，广东 广州 511400
  • 收稿日期:2022-09-06 出版日期:2022-12-20 发布日期:2022-12-20
  • 通讯作者: 叶沁丹
  • 作者简介:叶沁丹，浩鲸云计算科技股份有限公司，商发部副总经理/资深产品经理/市场经理/品牌经理，对BSS和OSS领域有深入的了解，现主要负责OSS支撑产品和算力网络产品的规划、市场和品牌。
    本文中负责论文整体架构编写及核心内容编写。
    YE Qindan, Deputy General Manager, as well as experienced Product Manager, Marketing Manager, and Brand Manager of the Commerce Development Department of Whale Cloud Technology Co., Ltd., has worked in the telecom operator business for 18 years, having an in-depth understanding of BSS and OSS fields. He is now responsible for products of the OSS support system and the planning, marketing, and brand promotion of computing network products.
    In this paper, he is responsible for the paper drafting and core content.
    E-mail: ye.qindan@iwhalecloud.com

Study on Integrated Support Solution and Application Scenarios of Computing Power Network

YE Qindan1,*(),FAN Guisheng2,Huang Hengyang1   

  1. 1. Whale Cloud Technology Co., Ltd, Business Development Department, Guangzhou, Guangdong 511400, China
    2. Whale Cloud Technology Co., Ltd, OSS Product Team, Guangzhou, Guangdong 511400, China
  • Received:2022-09-06 Online:2022-12-20 Published:2022-12-20
  • Contact: YE Qindan

RichHTML

0

PDF

169

摘要：

【目的】本文在“东数西算”国家工程背景下，从电信运营商的定位及挑战出发，重点介绍算力网络一体化支撑体系及应用场景的探索成果。【方法】本文从底层算力基础设施、中层算力编排管理层（算网大脑）、上层算力运营服务层三方面，阐述了算力网络意图受理、可信交易、统一纳管、统一度量、能力封装、编排调度、算方感知七大方面能力及多种落地场景，为算力供给方、需求方、运营方提供指导建议。【结果】电信运营商在算力网络发展历程中具备多重优势，通过体系化的构建，未来算力网络将在新媒体、云游戏、工业视觉、智慧家庭等丰富场景中落地。【局限】算力网络分三个阶段在不断演进，在技术路线、技术标准等多方面未形成统一，如何稳定推向市场商用还仍在试点，需于当前背景下在算网感知、算网运营、算网编排等核心能力方面不断提升加强，在不同场景中进行探索。【结论】在国家政策、技术发展双重驱动下，算力网络已成为新的发展方向，算力网络的发展将会上升到前所未有的高度。未来算力网络将成为新基建的一个核心的组成部分，渗透到生活的方方面面，加速驱动产业数字化转型。

关键词: 算力网络, 算网大脑, 算网编排, 算网调度

Abstract:

[Objective] The paper presents the challenges faced by the telecom operator as well as their orientation under the background of the national project of "East-West Computing Requirement Transfer". It expounds on the exploration of the integrated solution capability and application scenarios for the computing power network. [Methods] This paper mainly focuses on the seven aspects and application scenarios of the computing power network, including intent acceptance, trusted transaction, unified management, unified metrics, capacity encapsulation, orchestration scheduling, and computing party perception. The computing power network relies on the underlying computing infrastructure, the middle layer of computing power orchestration management (computing power network brain), and the upper layer of computing operation services. The paper aims to provide guidance and suggestions for computing power capability as well as the relevant operators. [Results] Telecom operators have the advantages in developing the computing power network. Through systematic solution construction, the future computing power network will be implemented in various scenarios such as social media, gaming on demand, industrial vision as well as smart home systems. [Limitations] The evolution of the computing power network can be divided into three stages to integrate and form the computing systems gradually. Works in computing ubiquity, computing power awareness, computing power routing, and computing power orchestration should be done on the basis of deep cloud-network integration. [Conclusions] As driven by both national policies and technology development, computing power network has shown a new development direction. The development of the computing power network will reach an unprecedented height. In the future, the computing power network will become a core component of the new infrastructure construction. It will penetrate into all aspects of people’s life and accelerate the digital transformation of the relevant industry.

Key words: computing power network, computing power network brain, computing power network orchestration, computing power network scheduling