数据与计算发展前沿 ›› 2025, Vol. 7 ›› Issue (2): 22-30.

CSTR: 32002.14.jfdc.CN10-1649/TP.2025.02.003

doi: 10.11871/jfdc.issn.2096-742X.2025.02.003

• 专刊：中国科技云10周年 • 上一篇    下一篇

可控热核聚变国际合作中网络加速的研究与实现

谭海波*()   

  1. 中国科学院合肥物质科学研究院，安徽 合肥 230031
  • 收稿日期:2025-02-12 出版日期:2025-04-20 发布日期:2025-04-23
  • 通讯作者: 谭海波
  • 作者简介:谭海波，中国科学院合肥物质科学研究院，博士，正高级工程师，信息中心主任，主要研究方向为网络数据传输性能优化、主动测量技术、区块链等。
    独立编写本文。
    TAN Haibo, Ph.D., is the Senior Engineer and director of the Information Center, Hefei Institutes of Physical Science, Chinese Academy of Sciences. His research interests cover network data transfer optimization, active network measurement, blockchain, etc.
    He accomplished the paper independently.
    E-mail: hbtan@hfcas.ac.cn

Research and Implementation of Network Acceleration in International Cooperation on Controlled Thermonuclear Fusion

TAN Haibo*()   

  1. Hefei Institutes of Physical Science, Chinese Academy of Sciences, Hefei, Anhui 230031, China
  • Received:2025-02-12 Online:2025-04-20 Published:2025-04-23
  • Contact: TAN Haibo

RichHTML

2

PDF

4

摘要：

【背景】可控热核聚变研究具有巨大政治、经济、社会意义；中国和全球主要国家在这个领域开展了全面、深入的合作，其中基于网络的科研协同成为了必然选择。【目的】针对热核聚变国际科研协同中不同应用的特点，提升长延时高带宽国际链路的传输性能，确保设计协同和在线实验的顺利开展。【方法】依托中国科技网，研究和应用多种网络数据传输加速手段，分别提升端到端应用、专用高性能传输节点间（DTNs Data Transfer Nodes）的网络传输性能。【结果】测试以及实践结果表明，本文的方法有效地支撑了热核聚变国际科研设计协同和在线实验的网络需求。

关键词: 网络加速, 拥塞控制算法, 数据包负载缓存技术

Abstract:

[Context] Controlled thermonuclear fusion research has great political, economic, and social significance. China and major countries in the world have carried out comprehensive and in-depth cooperation in this field and the network-based scientific research collaboration has become an inevitable choice. [Objective] This study is to enhance the transmission performance of long-latency, high-bandwidth international links in response to the characteristics of different applications in international thermonuclear fusion research collaboration, and to ensure the smooth conduct of design collaboration and online experiments. [Methods] Relying on the China Science and Technology Network, this study investigates and applies various network data transfer acceleration methods to improve the network transmission performance between end-to-end applications and between dedicated high-performance data transfer nodes (DTNs). [Results] The testing and practical results verify that the methods proposed in this paper effectively support the network requirements for international thermonuclear fusion research design collaboration and online experiments.

Key words: network acceleration, congestion control, payload caching