Pipeline Throughput Optimization

Pipeline Throughput Optimization is a computational framework within distributed deep learning that specifically enhances the rate of model training iterations by minimizing inter-stage idle times and balancing computational load across pipeline stages in partitioned neural network architectures.

31

Publications

1K

Citations

135

Authors

59

Institutions

Publications per year

2017–2026

27

Authors

135

  1. ZZ
    Zhiru Zhang

    Cornell University, United States

    3 publications · 55 citations

    3

    h-index

  2. SD
    Steve Dai

    Cornell University, United States

    3 publications · 55 citations

    3

    h-index

  3. MT
    Mingxing Tan

    Google (United States), United States

    2 publications · 37 citations

    2

    h-index

  4. ZL
    Zongqing Lu

    Pennsylvania State University, United States

    2 publications · 44 citations

    2

    h-index

  5. TL
    Thomas La Porta

    Pennsylvania State University, United States

    2 publications · 44 citations

    2

    h-index

1–5 of 135

Venues

  1. arXiv (Cornell University)

    Repository · 2 publications · 110 citations

    2

    h-index

  2. IEEE Transactions on Mobile Computing

    TMCCI · 2 publications · 44 citations

    2

    h-index

  3. IEEE Transactions on Parallel and Distributed Systems

    TPDS · 1 publications · 22 citations

    1

    h-index

  4. 1

    h-index

  5. Methods in molecular biology

    Book series · 1 publications · 11 citations

    1

    h-index

1–5