OpenMC： Towards Simplifying Programming for TianHe Supercomputers

Modern petascale and future exascale systems are massively heterogeneous architectures. Developing productive intra-node programming models is crucial toward addressing their programming challenge. We introduce a directive- based intra-node programming model, OpenMC, and show that this new model can...

Full description

Saved in:
Bibliographic Details
Published inJournal of computer science and technology Vol. 29; no. 3; pp. 532 - 546
Main Author 廖湘科 杨灿群 唐滔 易会战 王锋 吴强 薛京灵
Format Journal Article
LanguageEnglish
Published Boston Springer US 01.05.2014
Springer Nature B.V
Subjects
Accelerators
Analysis
Artificial Intelligence
Computer programming
Computer Science
Computer simulation
Data Structures and Information Theory
Hardware
Information Systems Applications (incl.Internet)
Mathematical models
Nodes
OpenMP
Programming
Prototyping
Regular Paper
Software
Software Engineering
Studies
Supercomputers
Tasks
Theory of Computation
Workers
任务并行
硬件资源
编程模型
计算机节点
计算节点
超大规模
超级计算机
supercomputer
programming model
heterogeneous
MIC
Online AccessGet full text

Cover

More Information
Summary:Modern petascale and future exascale systems are massively heterogeneous architectures. Developing productive intra-node programming models is crucial toward addressing their programming challenge. We introduce a directive- based intra-node programming model, OpenMC, and show that this new model can achieve ease of programming, high performance, and the degree of portability desired for heterogeneous nodes, especially those in TianHe supercomputers. While existing models are geared towards oifloading computations to accelerators （typically one）, OpenMC alms to more uniformly and adequately exploit the potential offered by multiple CPUs and accelerators in a compute node. OpenMC achieves this by providing a unified abstraction of hardware resources as workers and facilitating the exploitation of asynchronous task parallelism on the workers. We present an overview of OpenMC, a prototyping implementation, and results from some initial comparisons with OpenMP and hand-written code in developing six applications on two types of nodes from TianHe supercomputers.
Bibliography:11-2296/TP
Modern petascale and future exascale systems are massively heterogeneous architectures. Developing productive intra-node programming models is crucial toward addressing their programming challenge. We introduce a directive- based intra-node programming model, OpenMC, and show that this new model can achieve ease of programming, high performance, and the degree of portability desired for heterogeneous nodes, especially those in TianHe supercomputers. While existing models are geared towards oifloading computations to accelerators （typically one）, OpenMC alms to more uniformly and adequately exploit the potential offered by multiple CPUs and accelerators in a compute node. OpenMC achieves this by providing a unified abstraction of hardware resources as workers and facilitating the exploitation of asynchronous task parallelism on the workers. We present an overview of OpenMC, a prototyping implementation, and results from some initial comparisons with OpenMP and hand-written code in developing six applications on two types of nodes from TianHe supercomputers.
Xiang-Ke Liao, Member, ACM, IEEE, Can-Qun Yang, Tao Tang,Hui-Zhan Yi, Feng Wang, Member, CCF, ACM, Qiang Wu, Member, IEEE, and Jingling Xue, Senior Member, IEEE, Member, ACM（1School of Computer Science, National University of Defense Technology, Changsha 410073, China 2School of Computer Science and Engineering, University of New South Wales, Sydney, Australia）
supercomputer, programming model, heterogeneous, MIC
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:1000-9000
1860-4749
DOI:10.1007/s11390-014-1447-4