咨询与建议

限定检索结果

文献类型

  • 1 篇 期刊文献
  • 1 篇 会议

馆藏范围

  • 2 篇 电子文献
  • 0 种 纸本馆藏

日期分布

学科分类号

  • 2 篇 工学
    • 2 篇 计算机科学与技术...
    • 1 篇 电气工程
    • 1 篇 软件工程

主题

  • 2 篇 distributed deep...
  • 1 篇 near data proces...
  • 1 篇 runtime
  • 1 篇 parallel process...
  • 1 篇 hybrid paralleli...
  • 1 篇 task paralleliza...
  • 1 篇 data parallelism
  • 1 篇 transformers
  • 1 篇 pipeline paralle...
  • 1 篇 training distrib...
  • 1 篇 servers
  • 1 篇 computational mo...
  • 1 篇 computational st...
  • 1 篇 distributed syst...
  • 1 篇 costs
  • 1 篇 pipelines
  • 1 篇 graphics process...
  • 1 篇 artificial neura...
  • 1 篇 data models
  • 1 篇 privacy

机构

  • 1 篇 ngd syst inc irv...
  • 1 篇 uc irvine irvine...
  • 1 篇 natl univ def te...

作者

  • 1 篇 alves vladimir
  • 1 篇 torabzadehkashi ...
  • 1 篇 liu weijie
  • 1 篇 li dongsheng
  • 1 篇 bobarshad hossei...
  • 1 篇 ge keshi
  • 1 篇 lai zhiquan
  • 1 篇 heydarigorji ali
  • 1 篇 chou pai h.
  • 1 篇 rezaei siavash
  • 1 篇 lu xicheng
  • 1 篇 lu kai
  • 1 篇 li shengwei

语言

  • 2 篇 英文
检索条件"主题词=distributed deep neural network training"
2 条 记 录,以下是1-10 订阅
排序:
AutoPipe-H: A Heterogeneity-Aware Data-Paralleled Pipeline Approach on Commodity GPU Servers
收藏 引用
IEEE TRANSACTIONS ON COMPUTERS 2025年 第4期74卷 1196-1209页
作者: Liu, Weijie Lu, Kai Lai, Zhiquan Li, Shengwei Ge, Keshi Li, Dongsheng Lu, Xicheng Natl Univ Def Technol Coll Comp Changsha 410073 Hunan Peoples R China
Recently, the data-parallel pipeline approach has been widely used in training DNN models on commodity GPU servers. However, there are still three challenges for hybrid parallelism on commodity GPU servers: i) a balan... 详细信息
来源: 评论
Stannis: Low-Power Acceleration of DNN training Using Computational Storage Devices  57
Stannis: Low-Power Acceleration of DNN Training Using Comput...
收藏 引用
57th ACM/IEEE Design Automation Conference (DAC)
作者: HeydariGorji, Ali Torabzadehkashi, Mahdi Rezaei, Siavash Bobarshad, Hossein Alves, Vladimir Chou, Pai H. UC Irvine Irvine CA 92697 USA NGD Syst Inc Irvine CA USA
Computational storage devices enable in-storage processing of data in place. These devices contain 64-bit application processors and hardware accelerators that can help improving performance and saving power by reduci... 详细信息
来源: 评论