Efficient Large Message Broadcast using NCCL and CUDA-Aware MPI for Deep Learning
Author:
Affiliation:
1. Dept of Computer Science and Engineering, The Ohio State University
Publisher
ACM
Link
https://dl.acm.org/doi/pdf/10.1145/2966884.2966912
Reference26 articles.
1. KESCH : Cray CS-Storm System (CSCS). http://www.cscs.ch/computers/kesch_escha/index.html. KESCH: Cray CS-Storm System (CSCS). http://www.cscs.ch/computers/kesch_escha/index.html.
2. CNTK. http://www.cntk.ai/ 2015. CNTK. http://www.cntk.ai/ 2015.
3. Amazon. Deep Scalable Sparse Tensor Network Engine. https://github.com/amznlabs/amazon-dsstne 2016. Amazon. Deep Scalable Sparse Tensor Network Engine. https://github.com/amznlabs/amazon-dsstne 2016.
4. Interprocessor collective communication library (InterCom)
5. D. Bureddy H. Wang A. Venkatesh S. Potluri and D. Panda . OMB-GPU: A Micro-Benchmark Suite for Evaluating MPI Libraries on GPU Clusters. In J. Träff S. Benkner and J. Dongarra editors Recent Advances in the Message Passing Interface volume 7490 of Lecture Notes in Computer Science pages 110 -- 120 . Springer Berlin Heidelberg 2012 . ISBN 978-3-642-33517-4. URL http://dx.doi.org/10.1007/978-3-642-33518-1_16. 10.1007/978-3-642-33518-1_16 D. Bureddy H. Wang A. Venkatesh S. Potluri and D. Panda. OMB-GPU: A Micro-Benchmark Suite for Evaluating MPI Libraries on GPU Clusters. In J. Träff S. Benkner and J. Dongarra editors Recent Advances in the Message Passing Interface volume 7490 of Lecture Notes in Computer Science pages 110--120. Springer Berlin Heidelberg 2012. ISBN 978-3-642-33517-4. URL http://dx.doi.org/10.1007/978-3-642-33518-1_16. 10.1007/978-3-642-33518-1_16
Cited by 32 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
1. Distributed out-of-memory NMF on CPU/GPU architectures;The Journal of Supercomputing;2023-09-08
2. DeAR: Accelerating Distributed Deep Learning with Fine-Grained All-Reduce Pipelining;2023 IEEE 43rd International Conference on Distributed Computing Systems (ICDCS);2023-07
3. FMI: Fast and Cheap Message Passing for Serverless Functions;Proceedings of the 37th International Conference on Supercomputing;2023-06-21
4. Lyra: Elastic Scheduling for Deep Learning Clusters;Proceedings of the Eighteenth European Conference on Computer Systems;2023-05-08
5. The Performance of Distributed Applications: A Traffic Shaping Perspective;Proceedings of the 2023 ACM/SPEC International Conference on Performance Engineering;2023-04-15
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