Guardado en:
Detalles Bibliográficos
Autores principales: Li, Qi, Li, Kun, Han, Haozhi, Yuan, Liang, Chen, Junshi, Zhang, Yunquan, Chen, Yifeng, An, Hong, Cao, Ting, Yang, Mao
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:https://arxiv.org/abs/2506.22969
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866912455939063808
author Li, Qi
Li, Kun
Han, Haozhi
Yuan, Liang
Chen, Junshi
Zhang, Yunquan
Chen, Yifeng
An, Hong
Cao, Ting
Yang, Mao
author_facet Li, Qi
Li, Kun
Han, Haozhi
Yuan, Liang
Chen, Junshi
Zhang, Yunquan
Chen, Yifeng
An, Hong
Cao, Ting
Yang, Mao
contents Sparse Tensor Cores offer exceptional performance gains for AI workloads by exploiting structured 2:4 sparsity. However, their potential remains untapped for core scientific workloads such as stencil computations, which exhibit irregular sparsity patterns.This paper presents SparStencil, the first system to retarget sparse TCUs for scientific stencil computations through structured sparsity transformation. SparStencil introduces three key techniques: (1) Adaptive Layout Morphing, which restructures stencil patterns into staircase-aligned sparse matrices via a flatten-and-crush pipeline; (2) Structured Sparsity Conversion, which formulates transformation as a graph matching problem to ensure compatibility with 2:4 sparsity constraints; (3) Automatic Kernel Generation, which compiles transformed stencils into optimized sparse MMA kernels via layout search and table-driven memory mapping. Evaluated on 79 stencil kernels spanning diverse scientific domains, SparStencil achieves up to 7.1x speedup (3.1x on average) over state-of-the-art framework while reducing code complexity and matching or exceeding expert-tuned performance in both compute throughput and memory efficiency.
format Preprint
id arxiv_https___arxiv_org_abs_2506_22969
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SparStencil: Retargeting Sparse Tensor Cores to Scientific Stencil Computations via Structured Sparsity Transformation
Li, Qi
Li, Kun
Han, Haozhi
Yuan, Liang
Chen, Junshi
Zhang, Yunquan
Chen, Yifeng
An, Hong
Cao, Ting
Yang, Mao
Computational Engineering, Finance, and Science
Sparse Tensor Cores offer exceptional performance gains for AI workloads by exploiting structured 2:4 sparsity. However, their potential remains untapped for core scientific workloads such as stencil computations, which exhibit irregular sparsity patterns.This paper presents SparStencil, the first system to retarget sparse TCUs for scientific stencil computations through structured sparsity transformation. SparStencil introduces three key techniques: (1) Adaptive Layout Morphing, which restructures stencil patterns into staircase-aligned sparse matrices via a flatten-and-crush pipeline; (2) Structured Sparsity Conversion, which formulates transformation as a graph matching problem to ensure compatibility with 2:4 sparsity constraints; (3) Automatic Kernel Generation, which compiles transformed stencils into optimized sparse MMA kernels via layout search and table-driven memory mapping. Evaluated on 79 stencil kernels spanning diverse scientific domains, SparStencil achieves up to 7.1x speedup (3.1x on average) over state-of-the-art framework while reducing code complexity and matching or exceeding expert-tuned performance in both compute throughput and memory efficiency.
title SparStencil: Retargeting Sparse Tensor Cores to Scientific Stencil Computations via Structured Sparsity Transformation
topic Computational Engineering, Finance, and Science
url https://arxiv.org/abs/2506.22969