MMStencil accelerates 3D high-order stencils on a RISC multicore CPU's matrix units, claiming up to 2.1x over A100 GPU libraries and a 1.8x speedup for RTM seismic imaging.
Yask—yet another stencil kernel: A framework for hpc stencil code-generation and tuning,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.DC 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
MMStencil: Optimizing High-order Stencils on Multicore CPU using Matrix Unit
MMStencil accelerates 3D high-order stencils on a RISC multicore CPU's matrix units, claiming up to 2.1x over A100 GPU libraries and a 1.8x speedup for RTM seismic imaging.