XcalableACC: Extension of XcalableMP PGAS Language Using OpenACC for Accelerator Clusters

XcalableACC: Extension of XcalableMP PGAS Language Using OpenACC for Accelerator Clusters
复制标题

XcalableACC:使用 OpenACC 加速器集群的 XcalableMP PGAS 语言扩展

DOI:
--
复制
发表时间:
2014
期刊:
2014 First Workshop on Accelerator Programming using Directives
影响因子:
--
通讯作者:
M. Sato
M. Sato
中科院分区:
--
文献类型:
--
作者:
M. Nakao;H. Murai;Takenori Shimosaka;Akihiro Tabuchi;T. Hanawa;Yuetsu Kodama;T. Boku;M. Sato

文献摘要

参考文献

被引文献

相似文献

本文介绍了XcalableACC(XACC)编程模型,该模型是Xcalablemp(XMP)分区的全球地址空间(PGAS)语言和OpenACC的混合模型。 XACC定义了指令,使程序员能够混合XMP和OpenACC指令,以开发可以轻松使用加速器簇的应用程序。此外,为了提高模板应用的性能,Omni XACC编译器提供了可以通过紧密耦合加速器(TCA)在加速器内存上传输光环区域的功能,该网络是一个直接在加速器之间传输数据的网络。在本文中,我们通过实施Himeno基准来评估XACC的生产率和绩效。结果表明,由于提高了生产率,XACC需要少于代码线的一半与消息传递接口(MPI)和OpenACC的组合相比,通常将其一起用作典型的编程模型。由于这些性能的改进,使用TCA的XACC的性能最多要比通过Infiniband上GPUDIRECT RDMA的OpenACC和MPI编程模型的组合获得的速度要快2.7倍。
The present paper introduces the XcalableACC (XACC) programming model, which is a hybrid model of the XcalableMP (XMP) Partitioned Global Address Space (PGAS) language and OpenACC. XACC defines directives that enable programmers to mix XMP and OpenACC directives in order to develop applications that can use accelerator clusters with ease. Moreover, in order to improve the performance of stencil applications, the Omni XACC compiler provides functions that can transfer a halo region on accelerator memory via Tightly Coupled Accelerators (TCA), which is a proprietary network for transferring data directly among accelerators. In the present paper, we evaluate the productivity and the performance of XACC through implementations of the HIMENO Benchmark. The results show that thanks to the productivity improvements, XACC requires less than half the source lines of code compare to a combination of Message Passing Interface (MPI) and OpenACC, which is commonly used together as a typical programming model. As a result of these performance improvements, XACC using TCA achieved up to 2.7 times faster performance than could be obtained via the combination of OpenACC and MPI programming model using GPUDirect RDMA over InfiniBand.
DOI: 10.1177/1094342010391989
发表时间: 2011-02-01
影响因子: 3.1
作者:
Dongarra, Jack;Beckman, Pete;Yelick, Kathy
通讯作者: Yelick, Kathy
DOI: --
发表时间: 2009
期刊: Scientific Reports
影响因子: 4.6
作者:
J. Xu
通讯作者: J. Xu