Simt Execution Model
SIMT Execution Model is a parallel execution paradigm in GPU architecture where a single instruction stream is uniformly applied across multiple threads organized in warps, with individual threads diverging only through masking rather than separate instruction streams, thereby enabling efficient fine-grained parallelism.
34
Publications
938
Citations
100
Authors
40
Institutions
Publications per year
2017–2026
8
34
Neural acceleration for GPU throughput processors
Amir Yazdanbakhsh, Jongse Park, Hardik Sharma et al. · 2015 · 110 citations · Full text
Megakernels considered harmful
Samuli Laine, Tero Karras, Timo Aila · 2013 · 91 citations
The dual-path execution model for efficient GPU control flow
Minsoo Rhu, Mattan Erez · 2013 · 68 citations
Scheduling Page Table Walks for Irregular GPU Applications
Seunghee Shin, Guilherme Cox, Mark Oskin et al. · 2018 · 59 citations
Deploying Graph Algorithms on GPUs: An Adaptive Solution
Da Li, Michela Becchi · 2013 · 42 citations
1–5 of 34
100
4
h-index
4
h-index
3
h-index
3
h-index
3
h-index
1–5 of 100
40

Beihang University
Beijing, China
4 h-index
12 publications
297 citations

Xi'an Jiaotong University
Xi'an, China
4 h-index
9 publications
214 citations
Nvidia (United Kingdom)
Reading, United Kingdom
3 h-index
5 publications
334 citations
Northeastern University
Boston, United States
3 h-index
11 publications
135 citations
University of Wisconsin–Madison
Madison, United States
3 h-index
5 publications
127 citations
University of Florida
Gainesville, United States
3 h-index
6 publications
148 citations

University of British Columbia
Vancouver, Canada
3 h-index
4 publications
84 citations
Beijing Jiaotong University
Beijing, China
3 h-index
3 publications
83 citations
1–8 of 40
Conference · 1 publications · 12 citations
1
h-index
Repository · 1 publications · 14 citations
1
h-index
1 publications · 30 citations
1
h-index
MVR · 1 publications · 19 citations
1
h-index
Repository · 1 publications · 10 citations
1
h-index
1–5