RAJA: Portable Performance for Large-Scale Scientific Applications

2019 IEEE/ACM International Workshop on Performance, Portability and Productivity in HPC (P3HPC) Pub Date : 2019-11-01 DOI:10.1109/P3HPC49587.2019.00012

D. Beckingsale, T. Scogland, J. Burmark, R. Hornung, Holger E. Jones, W. Killian, A. Kunen, Olga Pearce, P. Robinson, B. Ryujin

{"title":"RAJA: Portable Performance for Large-Scale Scientific Applications","authors":"D. Beckingsale, T. Scogland, J. Burmark, R. Hornung, Holger E. Jones, W. Killian, A. Kunen, Olga Pearce, P. Robinson, B. Ryujin","doi":"10.1109/P3HPC49587.2019.00012","DOIUrl":null,"url":null,"abstract":"Modern high-performance computing systems are diverse, with hardware designs ranging from homogeneous multi- core CPUs to GPU or FPGA accelerated systems. Achieving desir- able application performance often requires choosing a program- ming model best suited to a particular platform. For large codes used daily in production that are under continual development, architecture-specific ports are untenable. Maintainability re- quires single-source application code that is performance portable across a range of architectures and programming models. In this paper we describe RAJA, a portability layer that enables C++ applications to leverage various programming models, and thus architectures, with a single-source codebase. We describe preliminary results using RAJA in three large production codes at Lawrence Livermore National Laboratory, observing 17×, 13× and 12× speedup on GPU-only over CPU- only nodes with single-source application code in each case.","PeriodicalId":377385,"journal":{"name":"2019 IEEE/ACM International Workshop on Performance, Portability and Productivity in HPC (P3HPC)","volume":"100 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"114","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2019 IEEE/ACM International Workshop on Performance, Portability and Productivity in HPC (P3HPC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/P3HPC49587.2019.00012","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 114

Abstract

Modern high-performance computing systems are diverse, with hardware designs ranging from homogeneous multi- core CPUs to GPU or FPGA accelerated systems. Achieving desir- able application performance often requires choosing a program- ming model best suited to a particular platform. For large codes used daily in production that are under continual development, architecture-specific ports are untenable. Maintainability re- quires single-source application code that is performance portable across a range of architectures and programming models. In this paper we describe RAJA, a portability layer that enables C++ applications to leverage various programming models, and thus architectures, with a single-source codebase. We describe preliminary results using RAJA in three large production codes at Lawrence Livermore National Laboratory, observing 17×, 13× and 12× speedup on GPU-only over CPU- only nodes with single-source application code in each case.

查看原文本刊更多论文

RAJA:大规模科学应用的便携性能

现代高性能计算系统是多种多样的，硬件设计范围从均匀的多核cpu到GPU或FPGA加速系统。实现理想的应用程序性能通常需要选择最适合特定平台的编程模型。对于在持续开发的生产环境中每天使用的大型代码，特定于体系结构的移植是站不住脚的。可维护性要求单源应用程序代码在各种体系结构和编程模型之间具有性能可移植性。在本文中，我们描述了RAJA，这是一个可移植性层，它使c++应用程序能够利用单一源代码库的各种编程模型和体系结构。我们描述了在劳伦斯利弗莫尔国家实验室的三个大型生产代码中使用RAJA的初步结果，在每种情况下，使用单源应用程序代码，仅gpu的节点比仅CPU的节点加速17倍，13倍和12倍。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

2019 IEEE/ACM International Workshop on Performance, Portability and Productivity in HPC (P3HPC)

自引率

0.00%

发文量