数据与计算发展前沿 ›› 2020, Vol. 2 ›› Issue (6): 1-10.doi: 10.11871/jfdc.issn.2096-742X.2020.06.001

• 专题:高性能计算在行业领域的特色应用 •    下一篇

2020年中国高性能计算机发展现状分析与展望

张云泉1,袁良1,*(),袁国兴2,李希代1   

  1. 1. 中国科学院计算技术研究所,计算机体系结构国家重点实验室,北京 100190
    2. 北京应用物理与计算数学研究所,北京 100088
  • 收稿日期:2020-11-23 出版日期:2020-12-20 发布日期:2020-12-29
  • 通讯作者: 袁良
  • 作者简介:张云泉,中国科学院计算技术研究所,计算机体系结构国家重点实验室,博士,研究员,博士生导师。 本文承担工作为:论文整体架构和布局,论文撰写。
    ZHANG Yunquan is a full professor and PhD Advisor of SKL of Computer Architecture, Institute of Computing Technology, Chinese Academy of Sciences in Beijing, China. He is also appointed as the Director of National Supercomputing Center at Jinan. His research interests are parallel algorithm and parallel software.He undertakes the following tasks in this paper : paper structure organizing and writting.E-mail:zyq@ict.ac.cn|袁良,中国科学院计算技术研究所,助理研究员,主要研究方向为并行计算模型、并行算法设计与程序优化。本文承担工作为:论文数据整理和分析。
    YUAN Liang is an assistant Professor of SKL of Computer Architecture, Institute of Computing Technology, Chinese Academy of Sciences in Beijing, China. His research interests are parallel computing model, parallel algorithm design and program optimization. He undertakes the following tasks in this paper: data arrangement and analysis. E-mail: yuanliang@ict.ac.cn|袁国兴,北京应用物理与计算数学研究所,研究员,研究方向为大规模科学与工程计算的方法与软件。本文承担工作为:论文布局。
    YUAN Guoxing is a research fellow at Institute of Applied Physics and Computational Mathematics, Chinese Academy of Sciences in Beijing, China. His research interests include large scale scientific and engineering computing methods and software development. He undertakes the following tasks in this paper: paper structure organizing.E-mail: ygx@iapcm.ac.cn|李希代,中国计算机学会高性能计算专委会,助理秘书长。本文承担工作为:数据收集与整理。
    LI Xidai is an assistant Secretary of TCHPC, CCF.She undertakes the following tasks in this paper: data collecting and arrangement.E-mail: lixidai@ict.ac.cn
  • 基金资助:
    国家重点研发计划(2016YFB0200803);国家重大科技基础设施项目“地球系统数值模拟装置”;国家自然科学基金(61972376);国家自然科学基金(62072431);国家自然科学基金(61432018);北京自然科学基金(L182053);中国科学院高效空间天气预报模式科技创新交叉与合作团队

State-of-the-Art Analysis and Perspectives of China HPC Development in 2020

ZHANG Yunquan1,YUAN Liang1,*(),YUAN Guoxing2,LI Xidai1   

  1. 1. State Key Laboratory of Computer Architecture, Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China
    2. Institute of Applied Physics and Computational Mathematics, Beijing 100088, China
  • Received:2020-11-23 Online:2020-12-20 Published:2020-12-29
  • Contact: YUAN Liang

摘要:

【目的】本文根据2020年11月发布的中国高性能计算机TOP100排行榜的数据,对国内高性能计算机的发展现状从总体性能、制造商、行业领域等方面进行了深入分析。【结果】中国TOP100的平均Linpack性能继续保持比国际TOP500平均Linpack性能高的局面,且TOP100的入门性能门槛仍然超过TOP500。中国TOP100上的超级计算系统出现了新面孔,Dell公司的一台机器上榜,打破了榜单连续两年全为国产超算系统的记录。联想为系统数量冠军,联想、曙光和浪潮三强争霸的局面继续保持。【结论】在此基础上,本文根据十九届排行榜的性能数据,对未来中国大陆高性能计算机的发展趋势进行了分析预测。根据新的数据,我们认为:峰值Exaflops的机器将在2021年到2022年间出现。

关键词: 高性能计算机, TOP100, 排行榜, 性能, 分析

Abstract:

[Objective] Based on the data of TOP100 China's high performance computer rankings published in November 2020, this paper makes an in-depth analysis of the current development status of high performance computers in China from the overall performance, manufacturers, industry status, and many other aspects. [Results] The average Linpack performance of TOP100 in China continues to be higher than those of the international TOP500, and the threshold for entry performance of TOP100 still exceeds that of TOP500. Dell built a new machine and makes itself a member in the TOP100 list, which broke the record that all systems in the list are domestic in recent two years. Lenovo is the champion in terms of number of systems in TOP100. The positions of the three strong hegemonies, Lenovo, Sugon, and Inspur, are unchanged and will be maintained and strengthened in the near future. [Conclusions] According to the performance data of the nineteenth ranking list, this paper analyzes and predicts the development trend of high-performance computers in China. Based on the new data, we believe that machines with Exaflops peak performance will appear between 2021 and 2022.

Key words: high performance computer, TOP100, ranking, performance, analysis