Next Article in Journal
An Evolutionary Computing Model for the Study of Within-Host Evolution
Next Article in Special Issue
ThunderX2 Performance and Energy-Efficiency for HPC Workloads
Previous Article in Journal
Heterogeneous Computing (CPU–GPU) for Pollution Dispersion in an Urban Environment
Article

GPU Computing with Python: Performance, Energy Efficiency and Usability

1
Mathematics and Cybernetics, SINTEF Digital, P.O. Box 124 Blindern, NO-0314 Oslo, Norway
2
Department of Mathematical Sciences, Norwegian University of Science and Technology, NO-7491 Trondheim, Norway
3
Research and Development Department, Norwegian Meteorological Institute, P.O. Box 43 Blindern, NO-0313 Oslo, Norway
4
Department of Computer Science, Oslo Metropolitan University, P.O. Box 4 St. Olavs plass, NO-0130 Oslo, Norway
5
Information Technology Department, Norwegian Meteorological Institute, P.O. Box 43 Blindern, NO-0313 Oslo, Norway
*
Author to whom correspondence should be addressed.
This paper is an extended version of our paper published in International Conference on Parallel Computing (ParCo2019).
Computation 2020, 8(1), 4; https://doi.org/10.3390/computation8010004
Received: 6 December 2019 / Revised: 30 December 2019 / Accepted: 6 January 2020 / Published: 9 January 2020
(This article belongs to the Special Issue Energy-Efficient Computing on Parallel Architectures)
In this work, we examine the performance, energy efficiency, and usability when using Python for developing high-performance computing codes running on the graphics processing unit (GPU). We investigate the portability of performance and energy efficiency between Compute Unified Device Architecture (CUDA) and Open Compute Language (OpenCL); between GPU generations; and between low-end, mid-range, and high-end GPUs. Our findings showed that the impact of using Python is negligible for our applications, and furthermore, CUDA and OpenCL applications tuned to an equivalent level can in many cases obtain the same computational performance. Our experiments showed that performance in general varies more between different GPUs than between using CUDA and OpenCL. We also show that tuning for performance is a good way of tuning for energy efficiency, but that specific tuning is needed to obtain optimal energy efficiency. View Full-Text
Keywords: GPU computing; CUDA; OpenCL; high-performance computing; shallow-water simulation; power efficiency GPU computing; CUDA; OpenCL; high-performance computing; shallow-water simulation; power efficiency
Show Figures

Figure 1

MDPI and ACS Style

Holm, H.H.; Brodtkorb, A.R.; Sætra, M.L. GPU Computing with Python: Performance, Energy Efficiency and Usability. Computation 2020, 8, 4. https://doi.org/10.3390/computation8010004

AMA Style

Holm HH, Brodtkorb AR, Sætra ML. GPU Computing with Python: Performance, Energy Efficiency and Usability. Computation. 2020; 8(1):4. https://doi.org/10.3390/computation8010004

Chicago/Turabian Style

Holm, Håvard H.; Brodtkorb, André R.; Sætra, Martin L. 2020. "GPU Computing with Python: Performance, Energy Efficiency and Usability" Computation 8, no. 1: 4. https://doi.org/10.3390/computation8010004

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Search more from Scilit
 
Search
Back to TopTop