Characterizing and Understanding GCNs on GPU

Characterizing and Understanding GCNs on GPU
复制标题

表征和理解 GPU 上的 GCN

DOI:
10.1109/lca.2020.2970395
复制
发表时间:
2020-01
影响因子:
2.3
通讯作者:
Yuan Xie
Yuan Xie
中科院分区:
计算机科学3区
文献类型:
--
作者:
Mingyu Yan;Zhaodong Chen;Lei Deng;Xiaochun Ye;Zhimin Zhang;Dongrui Fan;Yuan Xie

文献摘要

参考文献

被引文献

相似文献

图卷积神经网络(GCN)在图结构数据分析方面已经达到了最先进的性能。与传统的神经网络一样,GCN的训练和推理都是通过GPU来加速的。因此,表征和理解GCN在GPU上的执行模式对于软件和硬件优化都很重要。不幸的是,据我们所知,GPU上没有GCN工作负载的详细表征工作。在这封信中,我们将在推理阶段描述GCN工作负载,并探索NVIDIA V100 GPU上的GCN模型。给出了表征和探索,我们提出了几个有用的指导方针,软件优化和硬件优化的GCN在GPU上的有效执行。
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.
DOI: 10.1109/hpca.2019.00051
发表时间: 2019-02
期刊: 2019 IEEE International Symposium on High Performance Computer Architecture (HPCA)
影响因子: --
作者:
Abanti Basak;Shuangchen Li;Xing Hu;Sangmin Oh;Xinfeng Xie;Li Zhao;Xiaowei Jiang;Yuan Xie
通讯作者: Abanti Basak;Shuangchen Li;Xing Hu;Sangmin Oh;Xinfeng Xie;Li Zhao;Xiaowei Jiang;Yuan Xie
DOI: 10.1145/2851141.2851145
发表时间: 2016-08-01
影响因子: --
作者:
Wang, Yangzihao;Davidson, Andrew;Owens, John D.
通讯作者: Owens, John D.
DOI: 10.1126/science.4.99.760
发表时间: 1896-11
期刊: Science
影响因子: 56.9
作者:
E. D. Cope
通讯作者: E. D. Cope
DOI: 10.1145/3016078.2851145
发表时间: 2016-08-01
影响因子: --
作者:
Wang, Yangzihao;Davidson, Andrew;Owens, John D.
通讯作者: Owens, John D.
DOI: 10.14778/3352063.3352127
发表时间: 2019-02
期刊: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining
影响因子: --
作者:
Rong Zhu;Kun Zhao;Hongxia Yang;Wei Lin;Chang Zhou;Baole Ai;Yong Li;Jingren Zhou
通讯作者: Rong Zhu;Kun Zhao;Hongxia Yang;Wei Lin;Chang Zhou;Baole Ai;Yong Li;Jingren Zhou