Rubik: A Hierarchical Architecture for Efficient Graph Neural Network Training

doi:10.1109/TCAD.2021.3079142

	Rubik: A Hierarchical Architecture for Efficient Graph Neural Network Training
	Chen, Xiaobing 1,2; Wang, Yuke 3; Xie, Xinfeng 4; Hu, Xing 1; Basak, Abanti 4; Liang, Ling 4; Yan, Mingyu 1; Deng, Lei 5; Ding, Yufei 3; Du, Zidong 1; Xie, Yuan 4
	2022-04-01
发表期刊	IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS
ISSN	0278-0070
卷号	41 期号:4 页码:936-949
摘要	The graph convolutional network (GCN) emerges as a promising direction to learn the inductive representation in graph data commonly used in widespread applications, such as E-commerce, social networks, and knowledge graphs. However, learning from graphs is nontrivial because of its mixed computation model involving both graph analytics and neural network computing. To this end, we decompose the GCN learning into two hierarchical paradigms: 1) graph-level and 2) node-level computing. Such a hierarchical paradigm facilitates the software and hardware accelerations for GCN learning. We propose a lightweight graph reordering methodology, incorporated with a GCN accelerator architecture that equips a customized cache design to fully utilize the graph-level data reuse. We also propose a mapping methodology aware of data reuse and task-level parallelism to handle various graphs inputs effectively. The results show that Rubik accelerator design improves energy efficiency by 26.3x-1375.2x than GPU platforms across different datasets and GCN models.
关键词	Deep learning accelerator graph neural network (GNN)
DOI	10.1109/TCAD.2021.3079142
收录类别	SCI
语种	英语
资助项目	National Key Research and Development Program of China[2017YFA0700902] ; NSF of China[61925208] ; NSF of China[62002338] ; NSF of China[61732007] ; NSF of China[61732002] ; NSF of China[61906179] ; NSF of China[U19B2019] ; NSF of China[U20A20227] ; Beijing Natural Science Foundation[JQ18013] ; Strategic Priority Research Program of Chinese Academy of Science[XDB32050200] ; Strategic Priority Research Program of Chinese Academy of Science[XDC05010300] ; Beijing Academy of Artificial Intelligence (BAAI) ; Beijing Nova Program of Science and Technology[Z191100001119093] ; Youth Innovation Promotion Association CAS ; Xplore Prize
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Hardware & Architecture ; Computer Science, Interdisciplinary Applications ; Engineering, Electrical & Electronic
WOS记录号	WOS:000770597100014
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
引用统计	被引频次：18[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/18939
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Hu, Xing
作者单位	1.Chinese Acad Sci, Inst Comp Technol, State Key Lab Comp Architecture, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Sch Comp Sci & Technol, Beijing 100190, Peoples R China 3.Univ Calif Santa Barbara, Dept Comp Sci, Santa Barbara, CA 93106 USA 4.Univ Calif Santa Barbara, Dept Elect & Comp Engn, Santa Barbara, CA 93106 USA 5.Tsinghua Univ, Dept Precis Instrument, Ctr Brain Inspired Comp Res, Beijing 10084, Peoples R China
推荐引用方式 GB/T 7714	Chen, Xiaobing,Wang, Yuke,Xie, Xinfeng,et al. Rubik: A Hierarchical Architecture for Efficient Graph Neural Network Training[J]. IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS,2022,41(4):936-949.
APA	Chen, Xiaobing.,Wang, Yuke.,Xie, Xinfeng.,Hu, Xing.,Basak, Abanti.,...&Xie, Yuan.(2022).Rubik: A Hierarchical Architecture for Efficient Graph Neural Network Training.IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS,41(4),936-949.
MLA	Chen, Xiaobing,et al."Rubik: A Hierarchical Architecture for Efficient Graph Neural Network Training".IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS 41.4(2022):936-949.