Skip to main content

Research Repository

Advanced Search

GPU Fast Convolution via the Overlap-and-Save Method in Shared Memory

Adámek, Karel; Dimoudi, Sofia; Giles, Mike; Armour, Wesley

GPU Fast Convolution via the Overlap-and-Save Method in Shared Memory Thumbnail


Authors

Karel Adámek

Mike Giles

Wesley Armour



Abstract

We present an implementation of the overlap-and-save method, a method for the convolution of very long signals with short response functions, which is tailored to GPUs. We have implemented several FFT algorithms (using the CUDA programming language), which exploit GPU shared memory, allowing for GPU accelerated convolution. We compare our implementation with an implementation of the overlap-and-save algorithm utilizing the NVIDIA FFT library (cuFFT). We demonstrate that by using a shared-memory-based FFT, we can achieved significant speed-ups for certain problem sizes and lower the memory requirements of the overlap-and-save method on GPUs.

Citation

Adámek, K., Dimoudi, S., Giles, M., & Armour, W. (2020). GPU Fast Convolution via the Overlap-and-Save Method in Shared Memory. ACM Transactions on Architecture and Code Optimization, 17(3), Article 18. https://doi.org/10.1145/3394116

Journal Article Type Article
Acceptance Date Apr 30, 2020
Online Publication Date Aug 31, 2020
Publication Date 2020-08
Deposit Date Nov 18, 2020
Publicly Available Date Nov 18, 2020
Journal ACM Transactions on Architecture and Code Optimization
Print ISSN 1544-3566
Electronic ISSN 1544-3973
Publisher Association for Computing Machinery (ACM)
Peer Reviewed Peer Reviewed
Volume 17
Issue 3
Article Number 18
DOI https://doi.org/10.1145/3394116

Files







You might also like



Downloadable Citations