Synchronous Data Parallelism
Synchronous Data Parallelism is a data parallelism strategy in which all processing units simultaneously execute the same operation across different data partitions, coordinating their progress through global synchronization barriers at each computational step.
33
Publications
1.8K
Citations
124
Authors
50
Institutions
Publications per year
2017–2026
16
Parents
33
Yang You, Zhao Zhang, Cho‐Jui Hsieh et al. · 2018 · 408 citations
Shiqing Fan, Yi Rong, Meng Chen et al. · 2021 · 197 citations
CosmoFlow: Using Deep Learning to Learn the Universe at Scale
Amrita Mathuriya, Deborah Bard, P. J. Mendygral et al. · 2018 · 122 citations · Full text
Computing, Computer Science, Physics +17
1–5 of 33
124
3
h-index
3
h-index
3
h-index
2
h-index
2
h-index
1–5 of 124
50
University of California, Berkeley
Berkeley, United States
5 h-index
9 publications
1.5K citations
University of California, Davis
Davis, United States
2 h-index
2 publications
453 citations
Alibaba Group (United States)
San Mateo, United States
2 h-index
24 publications
2.7K citations

University of Hong Kong
Pok Fu Lam, Hong Kong
2 h-index
2 publications
226 citations
University of California San Diego
San Diego, United States
2 h-index
2 publications
67 citations
University of Washington
Seattle, United States
2 h-index
2 publications
67 citations
Berkeley College
New York, United States
2 h-index
6 publications
1.4K citations
Hong Kong Baptist University
Hong Kong, Hong Kong
2 h-index
7 publications
1K citations
1–8 of 50
Repository · 5 publications · 155 citations
5
h-index
1 publications · 35 citations
1
h-index
Conference · 1 publications · 18 citations
1
h-index
TACO · 1 publications · 10 citations
1
h-index
1 publications · 11 citations
1
h-index
1–5