HiHGNN: Accelerating HGNNs Through Parallelism and Data Reusability Exploitation

doi:10.1109/TPDS.2024.3394841

	HiHGNN: Accelerating HGNNs Through Parallelism and Data Reusability Exploitation
	Xue, Runzhen 1; Han, Dengke 1; Yan, Mingyu 1,2,3; Zou, Mo 4; Yang, Xiaocheng ; Wang, Duo 1; Li, Wenming 1; Tang, Zhimin 1; Kim, John 4; Ye, Xiaochun 1; Fan, Dongrui 1
	2024-07-01
发表期刊	IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS
ISSN	1045-9219
卷号	35 期号:7 页码:1122-1138
摘要	Heterogeneous graph neural networks (HGNNs) have emerged as powerful algorithms for processing heterogeneous graphs (HetGs), widely used in many critical fields. To capture both structural and semantic information in HetGs, HGNNs first aggregate the neighboring feature vectors for each vertex in each semantic graph and then fuse the aggregated results across all semantic graphs for each vertex. Unfortunately, existing graph neural network accelerators are ill-suited to accelerate HGNNs. This is because they fail to efficiently tackle the specific execution patterns and exploit the high-degree parallelism as well as data reusability inside and across the processing of semantic graphs in HGNNs. In this work, we first quantitatively characterize a set of representative HGNN models on GPU to disclose the execution bound of each stage, inter-semantic-graph parallelism, and inter-semantic-graph data reusability in HGNNs. Guided by our findings, we propose a high-performance HGNN accelerator, HiHGNN, to alleviate the execution bound and exploit the newfound parallelism and data reusability in HGNNs. Specifically, we first propose a bound-aware stage-fusion methodology that tailors to HGNN acceleration, to fuse and pipeline the execution stages being aware of their execution bounds. Second, we design an independency-aware parallel execution design to exploit the inter-semantic-graph parallelism. Finally, we present a similarity-aware execution scheduling to exploit the inter-semantic-graph data reusability. Compared to the state-of-the-art software framework running on NVIDIA GPU T4 and GPU A100, HiHGNN respectively achieves an average 40.0x and 8.3x speedup as well as 99.59% and 99.74% energy reduction with quintile the memory bandwidth of GPU A100.
关键词	Semantics Parallel processing Graph neural networks Vectors Graphics processing units Fuses Hardware GNN GNN accelerator graph neural network HGNN HGNN accelerator heterogeneous graph neural network
DOI	10.1109/TPDS.2024.3394841
收录类别	SCI
语种	英语
资助项目	National Key Research and Development Program of China
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Theory & Methods ; Engineering, Electrical & Electronic
WOS记录号	WOS:001224174400002
出版者	IEEE COMPUTER SOC
引用统计	被引频次：1[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/2XEOYT63/38955
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Yan, Mingyu
作者单位	1.Chinese Acad Sci, Inst Comp Technol, State Key Lab Processors, Beijing 100045, Peoples R China 2.Univ Chinese Acad Sci, Beijing 101408, Peoples R China 3.Shanghai Innovat Ctr Processor Technol, Shanghai 201203, Peoples R China 4.Korea Adv Inst Sci & Technol, Sch Elect Engn, Daejeon 305701, South Korea
推荐引用方式 GB/T 7714	Xue, Runzhen,Han, Dengke,Yan, Mingyu,et al. HiHGNN: Accelerating HGNNs Through Parallelism and Data Reusability Exploitation[J]. IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS,2024,35(7):1122-1138.
APA	Xue, Runzhen.,Han, Dengke.,Yan, Mingyu.,Zou, Mo.,Yang, Xiaocheng.,...&Fan, Dongrui.(2024).HiHGNN: Accelerating HGNNs Through Parallelism and Data Reusability Exploitation.IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS,35(7),1122-1138.
MLA	Xue, Runzhen,et al."HiHGNN: Accelerating HGNNs Through Parallelism and Data Reusability Exploitation".IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS 35.7(2024):1122-1138.