数据与计算发展前沿 ›› 2019, Vol. 1 ›› Issue (1): 11-21.doi: 10.11871/jfdc.issn.2096.742X.2019.01.003

• • 上一篇    下一篇

高性能计算框架软件——SC_Tangram

迟学斌1,2,*,赵莲1,*,王姗姗1,2,张鉴1,姜金荣1   

  1. 1.中国科学院计算机网络信息中心,北京 100190
    2.中国科学院大学,北京 100049
  • 收稿日期:2019-08-15 出版日期:2019-01-20 发布日期:2019-12-26
  • 通讯作者: 迟学斌,赵莲
  • 作者简介:迟学斌,1963年生,中国科学院计算机网络信息中心,研究员,博导,主要研究方向为高性能计算、并行计算。
    本文承担工作为:框架的整体结构设计、研究指导。
    Chi Xuebin, born in 1963, Computer Network Information Center of the Chinese Academy of Sciences, reasearch fellow, PhD supervisor. His main research interests are high performance computing and parallel computing.
    In this paper he undertakes the following tasks: the design of overall paper structure and research guidance of the frame.E-mail: chi@sccas.cn|赵莲,1989年生,中国科学院计算机网络信息中心,助理研究员,博士,主要研究方向为高性能计算、并行计算。
    本文承担工作为:框架代码实现,并行算法实现,力学应用测试。
    Zhao Lian, born in 1989, Computer Network Information Center of Chinese Academy of Sciences,assistant research fellow, PhD. Her main research interests are high performance computing and parallel computing.
    In this paper she undertakes the following tasks: code implementations of the whole framework, parallel algorithms and mechanical application testing.E-mail: zhaolian@sccas.cn|王姗姗,1995年生,中国科学院计算机网络信息中心,在读研究生,主要研究方向为高性能计算、相场模拟。
    本文承担工作为:相场模拟程序在框架上实现,应用测试。
    Wang Shanshan, born in 1995, Computer Network Information Center of Chinese Academy of Sciences,master student. Her main research interests are high performance computing and phase field simulation .
    In this paper she undertakes the following tasks: code implementations of the phase field simulation and its testing.E-mail:wangshanshan@cnic.cn|张鉴,1972年生,中国科学院计算机网络信息中心,研究员,博导,主要研究方向为高性能计算、偏微分方程数值算法和并行算法研究。
    本文承担工作为:相场模拟程序在框架上开发设计及指导。
    Zhang Jian, born in 1972, Computer Network Information Center of Chinese Academy of Sciences,research fellow, PhD supervisor. His main research interests are high performance computing, numerical algorithms for partial differential equations and parallel computing.
    In this paper he undertakes the following tasks: code design and execution director of the phase field simulation.E-mail:zhangjian@sccas.cn|姜金荣,1977年生,中国科学院计算机网络信息中心,研究员,主要研究方向为并行算法与框架软件、计算地球科学。
    本文承担工作为:框架代码开发指导。
    Jiang Jinrong, born in 1977, Computer Network Information Center of Chinese Academy of Sciences, research fellow. His main research interests are parallel computing algorithms and frameworks.
    In this paper he undertakes the following tasks: code design and execution director of the whole framework.E-mail:jjr@sccas.cn
  • 基金资助:
    国家重点研发计划“地球系统模式的改进、应用开发和高性能计算”(2016YFB0200800)

High Performance Computing Framework Software——SC_Tangram

Xuebin Chi1,2,*,Lian Zhao1,*,Shanshan Wang1,2,Jian Zhang1,Jinrong Jiang1   

  1. 1.Chinese Academy of Sciences, Computer Network Information Center, Beijing 100190, China
    2.University of Chinese Academy of Sciences, Beijing 100049, China
  • Received:2019-08-15 Online:2019-01-20 Published:2019-12-26
  • Contact: Xuebin Chi,Lian Zhao

摘要:

【目的】为降低并行编程难度,加速应用程序开发,本文设计并实现一种面向新型开发模式的并行框架软件——SC_Tangram,其中SC表示科学计算(Scientific Computing),Tangram(七巧板)寓意灵活组装。【方法】框架开发采用面向百亿亿次高性能计算的新型编程模型Charm++,为应用软件的并行扩展性和自适应性提供了保障。基于组件化软件开发方法,通过抽取应用中的共性部分,进行封装和隐藏,通过组件或配置文件接口的方式,供用户调用。【结果】针对现阶段的开发,框架已应用到力学计算、相场模拟等应用领域上,实验结果表明能得到较好的加速效果。【局限】目前框架软件上的功能模块还不全面,需针对不同应用需求开发相应的接口。【结论】SC_Tangram可以支持针对应用的共性和特性组件开发,随着在框架上开发更多的功能组件,未来将应用到更多的科学计算领域中。

关键词: 框架软件, Charm++, 组件化, 共性算法

Abstract:

[Objective]In order to reduce the difficulty of parallel programming and accelerate the development of application program , this paper designs and implements a parallel framework software, SC_Tangram, in which SC represents scientific computing and Tangram implies flexible assembly. [Methods]To guarantee the massively parallel scalability and adaptivity, the programming model Charm++ is adopted in the runtime system layer of the framework. By the method of component software development, SC_Tangram encapsulates and hides the common parts and can be invoked by users in term of component or configuration file interfaces. [Results]For the current development stage, the framework has been applied to mechanical calculation, phase field simulation and other applications. The experimental results show that it can perform more efficient computations. [Limitations]At present, the functional modules of the framework software are not comprehensive, so it is necessary to develop corresponding interfaces for different application requirements. [Conclusions]SC_Tangram can support the development of common and characteristic components for applications. With the development of more functional components in the framework, it will be applied to more fields of scientific computing in the future.

Key words: framework software, Charm++, componentization, common algorithms