A microbenchmark study maps memory hierarchy, execution pipelines, and FP4/FP6 tensor-core behavior on Nvidia's Blackwell RTX 5080 and compares it with Hopper's H100.
Characterizing and improving the use of demand-fetched caches in gpus,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.DC 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Dissecting the NVIDIA Blackwell Architecture with Microbenchmarks
A microbenchmark study maps memory hierarchy, execution pipelines, and FP4/FP6 tensor-core behavior on Nvidia's Blackwell RTX 5080 and compares it with Hopper's H100.