Register-Aware Optimizations for Parallel Sparse Matrix-Matrix Multiplication

doi:10.1007/s10766-018-0604-8

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

	Register-Aware Optimizations for Parallel Sparse Matrix-Matrix Multiplication
	He, Xin 2; Tan, Guangming 2; Liu, Junhong 2; Liu, Weifeng 1
刊名	INTERNATIONAL JOURNAL OF PARALLEL PROGRAMMING
	2019-06-01
卷号	47 期号:3 页码:403-417
关键词	Sparse matrix Sparse matrix-matrix multiplication GPU Register
ISSN号	0885-7458
DOI	10.1007/s10766-018-0604-8
英文摘要	General sparse matrix-matrix multiplication (SpGEMM) is a fundamental building block of a number of high-level algorithms and real-world applications. In recent years, several efficient SpGEMM algorithms have been proposed for many-core processors such as GPUs. However, their implementations of sparse accumulators, the core component of SpGEMM, mostly use low speed on-chip shared memory and global memory, and high speed registers are seriously underutilised. In this paper, we propose three novel register-aware SpGEMM algorithms for three representative sparse accumulators, i.e., sort, merge and hash, respectively. We fully utilise the GPU registers to fetch data, finish computations and store results out. In the experiments, our algorithms deliver excellent performance on a benchmark suite including 205 sparse matrices from the SuiteSparse Matrix Collection. Specifically, on an Nvidia Pascal P100 GPU, our three register-aware sparse accumulators achieve on average 2.0x (up to 5.4x), 2.6x (up to 10.5x) and 1.7x (up to 5.2x) speedups over their original implementations in libraries bhSPARSE, RMerge and NSPARSE, respectively.
资助项目	National Key Research and Development Program of China[2017YFB0202105] ; National Key Research and Development Program of China[2016YFB0201305] ; National Key Research and Development Program of China[2016YFB0200803] ; National Key Research and Development Program of China[2016YFB0200300] ; National Natural Science Foundation of China[61521092] ; National Natural Science Foundation of China[91430218] ; National Natural Science Foundation of China[31327901] ; National Natural Science Foundation of China[61472395] ; National Natural Science Foundation of China[61432018] ; European Union's Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie project[752321]
WOS研究方向	Computer Science
语种	英语
出版者	SPRINGER/PLENUM PUBLISHERS
WOS记录号	WOS:000471644400006
内容类型	期刊论文
源URL	[http://119.78.100.204/handle/2XEOYT63/4175]
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Liu, Junhong
作者单位	1.Norwegian Univ Sci & Technol, Dept Comp Sci, Trondheim, Norway 2.Univ Chinese Acad Sci, State Key Lab Comp Architecture, Inst Comp Technol, Chinese Acad Sci, Beijing, Peoples R China
推荐引用方式 GB/T 7714	He, Xin,Tan, Guangming,Liu, Junhong,et al. Register-Aware Optimizations for Parallel Sparse Matrix-Matrix Multiplication[J]. INTERNATIONAL JOURNAL OF PARALLEL PROGRAMMING,2019,47(3):403-417.
APA	He, Xin,Tan, Guangming,Liu, Junhong,&Liu, Weifeng.(2019).Register-Aware Optimizations for Parallel Sparse Matrix-Matrix Multiplication.INTERNATIONAL JOURNAL OF PARALLEL PROGRAMMING,47(3),403-417.
MLA	He, Xin,et al."Register-Aware Optimizations for Parallel Sparse Matrix-Matrix Multiplication".INTERNATIONAL JOURNAL OF PARALLEL PROGRAMMING 47.3(2019):403-417.