中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
Characterizing and Understanding GCNs on GPU

文献类型:期刊论文

作者Yan, Mingyu1,2,3; Chen, Zhaodong2; Deng, Lei2; Ye, Xiaochun; Zhang, Zhimin1; Fan, Dongrui1,3; Xie, Yuan1,2
刊名IEEE COMPUTER ARCHITECTURE LETTERS
出版日期2020
卷号19期号:1页码:22-25
关键词Graph convolutional neural networks characterization execution pattern GPU
ISSN号1556-6056
DOI10.1109/LCA.2020.2970395
英文摘要Graph convolutional neural networks (GCNs) have achieved state-of-the-art performance on graph-structured data analysis. Like traditional neural networks, training and inference of GCNs are accelerated with GPUs. Therefore, characterizing and understanding the execution pattern of GCNs on GPU is important for both software and hardware optimization. Unfortunately, to the best of our knowledge, there is no detailed characterization effort of GCN workloads on GPU. In this letter, we characterize GCN workloads at inference stage and explore GCN models on NVIDIA V100 GPU. Given the characterization and exploration, we propose several useful guidelines for both software optimization and hardware optimization for the efficient execution of GCNs on GPU.
资助项目National Natural Science Foundation of China[61732018] ; US National Science Foundation[1725447]
WOS研究方向Computer Science
语种英语
WOS记录号WOS:000525233900003
出版者IEEE COMPUTER SOC
源URL[http://119.78.100.204/handle/2XEOYT63/14266]  
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, ICT, SKLCA, Beijing 100864, Peoples R China
2.Univ Calif Santa Barbara, Santa Barbara, CA 93106 USA
3.UCAS, Beijing 100049, Peoples R China
推荐引用方式
GB/T 7714
Yan, Mingyu,Chen, Zhaodong,Deng, Lei,et al. Characterizing and Understanding GCNs on GPU[J]. IEEE COMPUTER ARCHITECTURE LETTERS,2020,19(1):22-25.
APA Yan, Mingyu.,Chen, Zhaodong.,Deng, Lei.,Ye, Xiaochun.,Zhang, Zhimin.,...&Xie, Yuan.(2020).Characterizing and Understanding GCNs on GPU.IEEE COMPUTER ARCHITECTURE LETTERS,19(1),22-25.
MLA Yan, Mingyu,et al."Characterizing and Understanding GCNs on GPU".IEEE COMPUTER ARCHITECTURE LETTERS 19.1(2020):22-25.

入库方式: OAI收割

来源:计算技术研究所

浏览0
下载0
收藏0
其他版本

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。