7301

On the Use of Small 2D Convolutions on GPUs

Shams A.H. Al Umairy, Alexander S. van Amesfoort, Irwan D. Setija, Martijn C. van Beurden, Henk J. Sips
Delft University of Technology, Delft, The Netherlands
Computer Architecture, Lecture Notes in Computer Science, Volume 6161/2012, 52-64, 2012

@inproceedings{al2012use,

   title={On the Use of Small 2D Convolutions on GPUs},

   author={Al Umairy, S. and van Amesfoort, A. and Setija, I. and van Beurden, M. and Sips, H.},

   booktitle={Computer Architecture},

   pages={52–64},

   year={2012},

   organization={Springer}

}

Download Download (PDF)   View View   Source Source   

673

views

Computing many small 2D convolutions using FFTs is a basis for a large number of applications in many domains in science and engineering, among them electromagnetic diffraction modeling in physics. The GPU architecture seems to be a suitable architecture to accelerate these convolutions, but reaching high application performance requires substantial development time and non-portable optimizations. In this work, we present the techniques, performance results and considerations to accelerate small 2D convolutions using CUDA, and compare performance to a multi-threaded CPU implementation. To improve programmability and performance of applications that make heavy use of small convolutions, we argue that two improvements to software and hardware are needed: FFT libraries must be extended with a single convolution function and communication bandwidth between CPU and GPU needs to be drastically improved.
No votes yet.
Please wait...

* * *

* * *

HGPU group © 2010-2017 hgpu.org

All rights belong to the respective authors

Contact us: