登录 注册 会员服务 联系我们

A GEMM interface and implementation on NVIDIA GPUs for multiple small matrices

  • Published:2015-01 Issue: Volume:75 Page:133-140
  • ISSN:0743-7315
  • Container-title:Journal of Parallel and Distributed Computing
  • language:en
  • Short-container-title:Journal of Parallel and Distributed Computing

Author:

Jhurani Chetan,Mullowney Paul

Funder

US Department of Energy SBIR

Publisher

Elsevier BV

Subject

Artificial Intelligence,Computer Networks and Communications,Hardware and Architecture,Theoretical Computer Science,Software

Reference13 articles.

1. LAPACK Users’ Guide;Anderson,1999

2. Sparse direct factorizations through unassembled hyper-matrices;Bientinesi;Comput. Methods Appl. Mech. Engrg.,2010

3. Computing with hp-Adaptive Finite Elements: Volume I: One and Two Dimensional Elliptic and Maxwell Problems;Demkowicz,2006

4. Computing with hp-Adaptive Finite Elements: Volume II Frontiers: Three Dimensional Elliptic and Maxwell Problems with Applications;Demkowicz,2007

5. High-Order Methods for Incompressible Fluid Flow;Deville,2002

Cited by 12 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. Multi-Level Encoding and Decoding in a Scalable Photonic Tensor Processor With a Photonic General Matrix Multiply (GeMM) Compiler;IEEE Journal of Selected Topics in Quantum Electronics;2022-11

2. Dissecting FLOPs Along Input Dimensions for GreenAI Cost Estimations;Machine Learning, Optimization, and Data Science;2022

3. Batched Small Tensor-Matrix Multiplications on GPUs;2020 IEEE 27th International Conference on High Performance Computing, Data, and Analytics (HiPC);2020-12

4. Low-Order Finite Element Solver with Small Matrix-Matrix Multiplication Accelerated by AI-Specific Hardware for Crustal Deformation Computation;Proceedings of the Platform for Advanced Scientific Computing Conference;2020-06-29

5. Enabling Highly Efficient Batched Matrix Multiplications on SW26010 Many-core Processor;ACM Transactions on Architecture and Code Optimization;2020-03-07
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