Asynchronous Data Parallelism

Asynchronous Data Parallelism is a data parallelism strategy in which model replicas independently perform computations on local data shards and update shared parameters without requiring synchronization across all replicas at each iteration, thereby allowing for varying update frequencies and reducing idle time in distributed training environments.

16

Publications

1.1K

Citations

76

Authors

27

Institutions

Publications per year

2017–2026

7

Authors

76

  1. BH
    Bingsheng He

    National University of Singapore, Singapore

    1 publications · 16 citations

    1

    h-index

  2. MP
    Manish Parashar

    Rutgers, The State University of New Jersey, United States

    1 publications · 137 citations

    1

    h-index

  3. JL
    Jiali Li

    Argonne National Laboratory, United States

    1 publications · 57 citations

    1

    h-index

  4. JM
    José María González-Linares

    Universidad de Málaga, Spain

    1 publications · 41 citations

    1

    h-index

  5. TA
    Tarek Abdelzaher

    University of Illinois Urbana-Champaign, United States

    1 publications · 11 citations

    1

    h-index

1–5 of 76

Venues

  1. IEEE Transactions on Nuclear Science

    1 publications · 14 citations

    1

    h-index

  2. Journal of Parallel and Distributed Computing

    1 publications · 41 citations

    1

    h-index

  3. Journal of Synchrotron Radiation

    1 publications · 40 citations

    1

    h-index

  4. IEEE Transactions on Software Engineering

    TSE · 1 publications · 110 citations

    1

    h-index

  5. Real-Time Systems

    1 publications · 11 citations

    1

    h-index

1–5