A multi-GPU parallel optimization model for the preconditioned conjugate gradient algorithm

doi:10.1016/j.parco.2017.04.003

CORC > 计算技术研究所 > 中国科学院计算技术研究所 > 中国科学院计算技术研究所期刊论文 > 英文

	A multi-GPU parallel optimization model for the preconditioned conjugate gradient algorithm
	Gao, Jiaquan 2,4; Zhou, Yuanshen 3; He, Guixia 1; Xia, Yifei 2
刊名	PARALLEL COMPUTING
	2017-04-01
卷号	63 页码:1-16
关键词	Optimization model Preconditioned conjugate gradient algorithm CUDA Multiple GPUs
ISSN号	0167-8191
DOI	10.1016/j.parco.2017.04.003
英文摘要	In this study, we present a novel optimization model that can automatically and rapidly generate an optimally parallel preconditioned conjugate gradient (PCG) algorithm for any given linear system on a specific multi-graphics processing unit (GPU) platform. For our proposed model, there are the following novelties: (1) a profile-based performance model for each one of the main components of the PCG algorithm, including the vector operation, inner product, and sparse matrix-vector multiplication (SpMV), is suggested, and (2) our model is general, independent of the problems, and only dependent on the resources of devices, and (3) our model is extensible. For a vector operation kernel, or inner product kernel, or SpMV kernel that is not included in our framework, once its performance model is successfully constructed, it can be incorporated into our framework. Our model is constructed only once for each type of GPU. The experiments validate the high efficiency of our proposed model. (C) 2017 Elsevier B.V. All rights reserved.
资助项目	Natural Science Foundation of Zhejiang Province, China[LY17F020021] ; Open Project Program of the State Key Laboratory of Computer Architecture[CARCH201603]
WOS研究方向	Computer Science
语种	英语
出版者	ELSEVIER SCIENCE BV
WOS记录号	WOS:000401212100001
内容类型	期刊论文
源URL	[http://119.78.100.204/handle/2XEOYT63/7182]
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Gao, Jiaquan
作者单位	1.Zhejiang Univ Technol, Zhijiang Coll, Hangzhou 310024, Zhejiang, Peoples R China 2.Nanjing Normal Univ, Sch Comp Sci & Technol, Nanjing 210023, Jiangsu, Peoples R China 3.Zhejiang Univ Technol, Coll Comp Sci & Technol, Hangzhou 310023, Zhejiang, Peoples R China 4.Chinese Acad Sci, Inst Comp Technol, State Key Lab Comp Architecture, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Gao, Jiaquan,Zhou, Yuanshen,He, Guixia,et al. A multi-GPU parallel optimization model for the preconditioned conjugate gradient algorithm[J]. PARALLEL COMPUTING,2017,63:1-16.
APA	Gao, Jiaquan,Zhou, Yuanshen,He, Guixia,&Xia, Yifei.(2017).A multi-GPU parallel optimization model for the preconditioned conjugate gradient algorithm.PARALLEL COMPUTING,63,1-16.
MLA	Gao, Jiaquan,et al."A multi-GPU parallel optimization model for the preconditioned conjugate gradient algorithm".PARALLEL COMPUTING 63(2017):1-16.