All-reduce Algorithms

All-Reduce Algorithms is a collective communication paradigm in cluster interconnects and cluster networking that belongs to the class of distributed synchronization methods, distinguished by its ability to perform a global reduction operation across all processing nodes and subsequently broadcast the result to every node in the system.

39

Publications

3.1K

Citations

142

Authors

51

Institutions

Publications per year

2017–2026

20

Authors

142

  1. WG
    William Gropp

    Argonne National Laboratory, United States

    4 publications · 1K citations

    4

    h-index

  2. DK
    Dhabaleswar K. Panda

    The Ohio State University, United States

    3 publications · 86 citations

    3

    h-index

  3. YT
    Yoshiki Tanaka

    Nara Institute of Science and Technology, Japan

    2 publications · 130 citations

    2

    h-index

  4. HS
    Hisahiro Suganuma

    2 publications · 130 citations

    2

    h-index

  5. PS
    Paul Sack

    University of Illinois Urbana-Champaign, United States

    2 publications · 45 citations

    2

    h-index

1–5 of 142

Venues

  1. arXiv (Cornell University)

    Repository · 4 publications · 496 citations

    4

    h-index

  2. IEEE Computer Architecture Letters

    1 publications · 24 citations

    1

    h-index

  3. IBM Journal of Research and Development

    1 publications · 87 citations

    1

    h-index

  4. Journal of Parallel and Distributed Computing

    1 publications · 397 citations

    1

    h-index

  5. 1

    h-index

1–5