Towards portable realizations of winograd-based convolution with vector intrinsics and OpenMP
Impact
![Google Scholar](/xmlui/themes/Mirage2/images/uji/logo_google.png)
![Microsoft Academico](/xmlui/themes/Mirage2/images/uji/logo_microsoft.png)
Metadata
Show full item recordcomunitat-uji-handle:10234/9
comunitat-uji-handle2:10234/7036
comunitat-uji-handle3:10234/146069
comunitat-uji-handle4:
INVESTIGACIONThis resource is restricted
https://doi.org/ 10.1109/PDP55904.2022.00015 |
Metadata
Title
Towards portable realizations of winograd-based convolution with vector intrinsics and OpenMPDate
2022Publisher
IEEEISBN
9781665469586Bibliographic citation
M. F. Dolz, A. Castelló and E. S. Quintana-Ortí, "Towards Portable Realizations of Winograd-based Convolution with Vector Intrinsics and OpenMP," 2022 30th Euromicro International Conference on Parallel, Distributed and Network-based Processing (PDP), 2022, pp. 39-46, doi: 10.1109/PDP55904.2022.00015.Type
info:eu-repo/semantics/conferenceObjectPublisher version
https://ieeexplore.ieee.org/document/9756700Version
info:eu-repo/semantics/publishedVersionSubject
Abstract
We take a step forward in the direction of developing high performance codes for the convolution, based on the Winograd transformation, that are easy to customize for different processor architectures. In our approach, ... [+]
We take a step forward in the direction of developing high performance codes for the convolution, based on the Winograd transformation, that are easy to customize for different processor architectures. In our approach, augmenting the portability of the solution is achieved via the introduction of vector intrinsics to exploit the SIMD (single-instruction multiple-data) capabilities of current processors as well as OpenMP pragmas to exploit multi-thread parallelism. While this comes at the cost of sacrificing a fraction of the computational performance, our experimental results on two distinct processors, with Intel Xeon Skylake and ARM Cortex A57 architectures, show that the impact is affordable, and still renders a Winograd-based solution that is competitive with the general method for the convolution based on the so-called im2col transform followed by a matrix-matrix multiplication.
URL: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9756700&isnumber=9756614 [-]
Description
Ponència presentada en el 2022 30th Euromicro International Conference on Parallel, Distributed and Network-based Processing (PDP)
Is part of
Actas 2022 30th Euromicro International Conference on Parallel, Distributed and Network-based Processing (PDP)Funder Name
Agencia Estatal de Investigacion | Generalitat Valenciana | Ministerio de Ciencia, Innovación y Universidades
Project code
PID2020- 113656RB-C21/C22 | Prometeo/2019/109 | FJC2019-039222-I | CDEIGENT/2018/014