CommBench: Micro-Benchmarking Hierarchical Networks with Multi-GPU, Multi-NIC Nodes

Author:

Hidayetoglu Mert1ORCID,De Gonzalo Simon Garcia2ORCID,Slaughter Elliott3ORCID,Li Yu4ORCID,Zimmer Christopher5ORCID,Bicer Tekin6ORCID,Ren Bin7ORCID,Gropp William8ORCID,Hwu Wen-Mei9ORCID,Aiken Alex1ORCID

Affiliation:

1. Stanford University, USA

2. Sandia National Laboratories, United States of America

3. SLAC National Accelerator Laboratory, USA

4. University of Illinois at Urbana-Champaign, USA

5. Oak Ridge National Laboratory, United States of America

6. Argonne National Laboratory, United States of America

7. William & Mary, USA

8. University of Illinois at Urbana-Champaign, United States of America

9. Nvidia Research, USA and University of Illinois at Urbana-Champaign, USA

Funder

Department of Energy

Publisher

ACM

Reference38 articles.

1. [1] [n. d.]. https://mvapich.cse.ohio-state.edu/benchmarks/

2. An application based MPI message throughput benchmark

3. Reducing communication in algebraic multigrid with multi-step node aware communication

4. Modeling Data Movement Performance on Heterogeneous Architectures

5. National Energy Research Scientific Computing Center. 2022. Perlmutter Architecture. https://docs.nersc.gov/systems/perlmutter/architecture/

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

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3