Accelerating Graph Processing With Lightweight Learning-Based Data Reordering

doi:10.1109/LCA.2022.3151087

	Accelerating Graph Processing With Lightweight Learning-Based Data Reordering
	Zou, Mo 1,2; Zhang, Mingzhe 3; Wang, Rujia 4; Sun, Xian-He 4; Ye, Xiaochun 1; Fan, Dongrui 1; Tang, Zhimin 1
	2022
发表期刊	IEEE COMPUTER ARCHITECTURE LETTERS
ISSN	1556-6056
卷号	21 期号:1 页码:5-8
摘要	Graph processing is a vital component in various application domains. However, a good graph processing performance is hard to achieve due to its intensive irregular data accesses. Noticing that in real-world graphs, a small portion of vertices occupy most connections, several techniques are proposed to reorder vertices based on their access frequency for better data access locality. However, these approaches can be further improved by identifying reordered data more effectively, which will reduce reordering overhead and improve overall performance. In this letter, we propose Learning-Based Reordering (LBR), a novel lightweight framework that identifies and reorders hot data adaptively for given graphs, algorithms, and threads. Our experimental evaluation indicates that LBR decreases reordering overhead by 24.7% while improves performance by 9.9% compared to the best-performing existing scheme.
关键词	Instruction sets Computational modeling Predictive models Big Data applications Software Optimization Graph processing reordering technique learning-based prediction model
DOI	10.1109/LCA.2022.3151087
收录类别	SCI
语种	英语
资助项目	National Science Foundation[CCF-2008907] ; National Science Foundation[CCF-2029014] ; National Natural Science Foundation of China[61732018] ; National Natural Science Foundation of China[61872335] ; National Natural Science Foundation of China[61802367] ; Austrian-Chinese Cooperative RD Project (FFG)[171111KYSB20200002] ; Austrian-Chinese Cooperative RD Project (CAS)[171111KYSB20200002]
WOS研究方向	Computer Science
WOS类目	Computer Science, Hardware & Architecture
WOS记录号	WOS:000764757400001
出版者	IEEE COMPUTER SOC
引用统计	被引频次：2[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/2XEOYT63/18951
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Zou, Mo
作者单位	1.Chinese Acad Sci, Inst Comp Technol, Beijing 100045, Peoples R China 2.Univ Chinese Acad Sci, Beijing 101408, Peoples R China 3.Chinese Acad Sci, Inst Informat Engn, Beijing 100045, Peoples R China 4.IIT, Chicago, IL 60616 USA
推荐引用方式 GB/T 7714	Zou, Mo,Zhang, Mingzhe,Wang, Rujia,et al. Accelerating Graph Processing With Lightweight Learning-Based Data Reordering[J]. IEEE COMPUTER ARCHITECTURE LETTERS,2022,21(1):5-8.
APA	Zou, Mo.,Zhang, Mingzhe.,Wang, Rujia.,Sun, Xian-He.,Ye, Xiaochun.,...&Tang, Zhimin.(2022).Accelerating Graph Processing With Lightweight Learning-Based Data Reordering.IEEE COMPUTER ARCHITECTURE LETTERS,21(1),5-8.
MLA	Zou, Mo,et al."Accelerating Graph Processing With Lightweight Learning-Based Data Reordering".IEEE COMPUTER ARCHITECTURE LETTERS 21.1(2022):5-8.