29241

Fast and Practical Strassen’s Matrix Multiplication using FPGAs

Afzal Ahmad, Linfeng Du, Wei Zhang
Department of Electronics and Computer Engineering, The Hong Kong University of Science and Technology, Hong Kong
arXiv:2406.02088 [cs.AR], (arXiv:2406.02088 [cs.AR])

@misc{ahmad2024fast,

   title={Fast and Practical Strassen’s Matrix Multiplication using FPGAs},

   author={Afzal Ahmad and Linfeng Du and Wei Zhang},

   year={2024},

   eprint={2406.02088},

   archivePrefix={arXiv},

   primaryClass={cs.AR}

}

Matrix multiplication is a cornerstone operation in a wide array of scientific fields, including machine learning and computer graphics. The standard algorithm for matrix multiplication has a complexity of O(n3) for n×n matrices. Strassen’s algorithm improves this to O(n2.807), but its practicality is limited for small to medium matrix sizes due to the large number of additions it introduces. This paper presents a novel FPGA-based implementation of Strassen’s algorithm that achieves superior speed over an optimized General Matrix Multiply (GeMM) implementation for matrices as small as n=256. Our design, tested extensively on two high-performance FPGA accelerators (Alveo U50 and U280) across various data types, matches or surpasses the performance of a highly optimized baseline across a range of matrix sizes.
No votes yet.
Please wait...

* * *

* * *

HGPU group © 2010-2024 hgpu.org

All rights belong to the respective authors

Contact us: