EUVIMEDEUVIMED entdecken
Crossref

Optimizing Structured-Sparse Matrix Multiplication in RISC-V Vector Processors

Vasileios Titopoulos, Kosmas Alexandridis, Christodoulos Peltekis, Chrysostomos Nicopoulos, Giorgos Dimitrakopoulos

IEEE Transactions on Computers · 2025-04 · 74 (4) · 1446-1460

DOI: 10.1109/tc.2025.3533083 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Computers, 2025-04. Verfasst von Vasileios Titopoulos und 4 weiteren Autor:innen. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 45 Literaturangaben. Gefördert von: Codasip.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Structured sparsity has been proposed as an efficient way to prune the complexity of Machine Learning (ML) applications and to simplify the handling of sparse data in hardware. Accelerating ML models, whether for training, or inference, heavily relies on matrix multiplications that can be efficiently executed on vector processors, or custom matrix engines. This work aims to integrate the simplicity of structured sparsity into vector execution to speed up the corresponding matrix multiplications. Initially, the implementation of structured-sparse matrix multiplication using the current RISC-V instruction set vector extension is comprehensively explored. Critical parameters that affect performance, such as the impact of data distribution across the scalar and vector register files, data locality, and the effectiveness of loop unrolling are analyzed both qualitatively and quantitatively. Furthermore, it is demonstrated that the addition of a single new instruction would reap even higher performance. The newly proposed instruction is called vindexmac, i.e., vector index-multiply-accumulate. It allows for indirect reads from the vector register file and it reduces the number of instructions executed per matrix multiplication iteration, without introducing additional dependencies that would limit loop unrolling. The proposed new instruction was integrated in a decoupled RISC-V vector processor with negligible hardware cost. Experimental results demonstrate the runtime efficiency and the scalability offered by the introduced optimizations and the new instruction for the execution of state-of-the-art Convolutional Neural Networks. More particularly, the addition of a custom instruction improves runtime by 25% and 33%, when compared with highly-optimized vectorized kernels that use only the currently defined RISC-V instructions.

Förderung

  • Codasip

Zitieren

APA

Titopoulos, V., Alexandridis, K., Peltekis, C., Nicopoulos, C., & Dimitrakopoulos, G. (2025). Optimizing Structured-Sparse Matrix Multiplication in RISC-V Vector Processors. IEEE Transactions on Computers, 74(4), 1446-1460. https://doi.org/10.1109/tc.2025.3533083

Vancouver

Titopoulos V, Alexandridis K, Peltekis C, Nicopoulos C, Dimitrakopoulos G. Optimizing Structured-Sparse Matrix Multiplication in RISC-V Vector Processors. IEEE Transactions on Computers. 2025;74(4):1446-1460. doi:10.1109/tc.2025.3533083

BibTeX
@article{Titopoulos2025,
  author = {Titopoulos, Vasileios and Alexandridis, Kosmas and Peltekis, Christodoulos and Nicopoulos, Chrysostomos and Dimitrakopoulos, Giorgos},
  title = {Optimizing Structured-Sparse Matrix Multiplication in RISC-V Vector Processors},
  journal = {IEEE Transactions on Computers},
  year = {2025},
  volume = {74},
  number = {4},
  pages = {1446--1460},
  doi = {10.1109/tc.2025.3533083}
}
Kostenlos starten →