623 Tflop/s HPCG run on Tianhe-2: Leveraging millions of hybrid cores

ISCAS OpenIR

	623 Tflop/s HPCG run on Tianhe-2: Leveraging millions of hybrid cores
	Liu, YQ; Yang, C; Liu, FF; Zhang, XY; Lu, YT; Du, YF; Yang, CQ; Xie, M; Liao, XK
	2016
发表期刊	INTERNATIONAL JOURNAL OF HIGH PERFORMANCE COMPUTING APPLICATIONS
ISSN	1094-3420
卷号	30 期号:1 页码:39-54
摘要	In this article, we present a new hybrid algorithm to enable and scale the high-performance conjugate gradients (HPCG) benchmark on large-scale heterogeneous systems such as the Tianhe-2. Based on an inner-outer subdomain partitioning strategy, the data distribution between host and device can be balanced adaptively. The overhead of data movement from both the MPI communication and the PCI-E transfer can be significantly reduced by carefully rearranging and fusing operations. A variety of parallelization and optimization techniques for performance-critical kernels are exploited and analyzed to maximize the performance gain on both host and device. We carry out experiments on both a small heterogeneous computer and the world's largest one, the Tianhe-2. On the small system, a thorough comparison and analysis has been presented to select from different optimization choices. On Tianhe-2, the optimized implementation scales to the full-system level of 3.12 million heterogeneous cores, with an aggregated performance of 623 Tflop/s and a parallel efficiency of 81.2%.; In this article, we present a new hybrid algorithm to enable and scale the high-performance conjugate gradients (HPCG) benchmark on large-scale heterogeneous systems such as the Tianhe-2. Based on an inner-outer subdomain partitioning strategy, the data distribution between host and device can be balanced adaptively. The overhead of data movement from both the MPI communication and the PCI-E transfer can be significantly reduced by carefully rearranging and fusing operations. A variety of parallelization and optimization techniques for performance-critical kernels are exploited and analyzed to maximize the performance gain on both host and device. We carry out experiments on both a small heterogeneous computer and the world's largest one, the Tianhe-2. On the small system, a thorough comparison and analysis has been presented to select from different optimization choices. On Tianhe-2, the optimized implementation scales to the full-system level of 3.12 million heterogeneous cores, with an aggregated performance of 623 Tflop/s and a parallel efficiency of 81.2%.
收录类别	SCI
关键词	Tianhe-2 Hpcg Conjugate Gradients Mic Heterogeneous Computing
部门归属	Chinese Acad Sci, Inst Software, Beijing, Peoples R China. Univ Chinese Acad Sci, Beijing, Peoples R China. Chinese Acad Sci, State Key Lab Comp Sci, Beijing, Peoples R China. Natl Univ Def Technol, Dept Comp Sci & Technol, Changsha, Hunan, Peoples R China.
语种	英语
WOS记录号	WOS:000371326000004
引用统计
内容类型	期刊论文
URI标识	http://ir.iscas.ac.cn/handle/311060/17346
专题	中国科学院软件研究所
推荐引用方式 GB/T 7714	Liu, YQ,Yang, C,Liu, FF,et al. 623 Tflop/s HPCG run on Tianhe-2: Leveraging millions of hybrid cores[J]. INTERNATIONAL JOURNAL OF HIGH PERFORMANCE COMPUTING APPLICATIONS,2016,30(1):39-54.
APA	Liu, YQ.,Yang, C.,Liu, FF.,Zhang, XY.,Lu, YT.,...&Liao, XK.(2016).623 Tflop/s HPCG run on Tianhe-2: Leveraging millions of hybrid cores.INTERNATIONAL JOURNAL OF HIGH PERFORMANCE COMPUTING APPLICATIONS,30(1),39-54.
MLA	Liu, YQ,et al."623 Tflop/s HPCG run on Tianhe-2: Leveraging millions of hybrid cores".INTERNATIONAL JOURNAL OF HIGH PERFORMANCE COMPUTING APPLICATIONS 30.1(2016):39-54.