A Memory-Aware Sparse Matrix-Matrix Multiplication on Multicore Architectures
ACM Transactions on Architecture and Code Optimization · 30.06.2026 · 23 (2) · 1-21
DOI: 10.1145/3803422 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACM Transactions on Architecture and Code Optimization, online erschienen am 26.06.2026. Verfasst von Deshun Bi und 8 weiteren Autor:innen. Die Arbeit nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: National Key R&D Program of China, National Natural Science Foundation of China.
Abstract (Originalsprache)
Sparse matrix–matrix multiplication (SpMM) is a fundamental operation in scientific computing with broad applications across numerous domains. Tiling is a key optimization technique for improving data locality and is widely adopted in high-performance computing. However, the irregular data access patterns inherent to SpMM make it challenging to exploit tiling effectively for data reuse. In this article, we propose MaSpMM , a memory-aware SpMM framework that integrates cache-aware tiling with a segment-oriented data layout. MaSpMM stores matrices as continuous segments to enhance data locality within each tile. Moreover, since many sparse matrices in real-world applications exhibit symmetry, we further develop MaSpMM-Sym, an extension that recursively partitions symmetric matrices to eliminate write conflicts and further improve locality. To adapt to diverse scenarios, we finally introduce MaSpMM-Adap, which adaptively selects the most suitable approach for each input matrix. Comprehensive evaluations on both x86 and ARM CPUs demonstrate that MaSpMM-Adap achieves average speedups of up to 1.86× over Intel oneMKL, 1.84× over ASpT, and 1.75× over J-Stream.
Förderung
- National Key R&D Program of China
- National Natural Science Foundation of China
Zitieren
APA
Bi, D., Li, S., Qiu, H., Xu, C., Yang, X., Yuan, F., Dong, D., Xiao, T., & Liu, J. (2026). A Memory-Aware Sparse Matrix-Matrix Multiplication on Multicore Architectures. ACM Transactions on Architecture and Code Optimization, 23(2), 1-21. https://doi.org/10.1145/3803422
Vancouver
Bi D, Li S, Qiu H, Xu C, Yang X, Yuan F, et al. A Memory-Aware Sparse Matrix-Matrix Multiplication on Multicore Architectures. ACM Transactions on Architecture and Code Optimization. 2026;23(2):1-21. doi:10.1145/3803422
BibTeX
@article{Bi2026,
author = {Bi, Deshun and Li, Shengguo and Qiu, Haozhong and Xu, Chuanfu and Yang, Xiaojian and Yuan, Fan and Dong, Dezun and Xiao, Tiaojie and Liu, Jie},
title = {A Memory-Aware Sparse Matrix-Matrix Multiplication on Multicore Architectures},
journal = {ACM Transactions on Architecture and Code Optimization},
year = {2026},
volume = {23},
number = {2},
pages = {1--21},
doi = {10.1145/3803422}
}
EUVIMED