Sparsity-Aware Communication for Distributed Graph Neural Network Training-Reference-Cited by-同舟云学术

登录注册会员服务联系我们

Sparsity-Aware Communication for Distributed Graph Neural Network Training

Published:2024-08-12 Issue: Volume:30 Page:117-126
ISSN:
Container-title:Proceedings of the 53rd International Conference on Parallel Processing
language:
Short-container-title:

Author:

Mukhopadhyay Ujjaini¹^ORCID,Tripathy Alok²^ORCID,Selvitopi Oguz³^ORCID,Yelick Katherine²^ORCID,Buluc Aydin⁴^ORCID

Affiliation:

1. University of California, Berkeley, United States of America

2. University of California, Berkeley, United States of America and Lawrence Berkeley National Lab, United States of America

3. Lawrence Berkeley National Lab, USA

4. Lawrence Berkeley National Lab, USA and University of California, Berkeley, USA

Funder

DOE U.S. Department of Energy

Publisher

ACM

Link

https://dl.acm.org/doi/pdf/10.1145/3673038.3673152

Reference30 articles.

1. Computing Graph Neural Networks: A Survey from Algorithms to Accelerators

2. Improving performance of sparse matrix dense matrix multiplication on large-scale parallel systems

3. Kadir Akbudak, Oguz Selvitopi, and Cevdet Aykanat. 2018. Partitioning Models for Scaling Parallel Sparse Matrix-Matrix Multiplication. TOPC 4, 3, Article 13 (2018), 34 pages.

4. Exploiting Multiple Levels of Parallelism in Sparse Matrix-Matrix Multiplication

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线，采集、加工和组织学术论文而形成的新型学术文献查询和分析系统，可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容，当前同舟云学术共收录了国内外主流学术期刊6万余种，收集的期刊论文及会议论文总量共计约1.5亿篇，并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询！咨询电话：010-8811{复制后删除}0370

www.globalauthorid.com

TOP