Contract-based general-purpose GPU programming

Using GPUs as general-purpose processors has revolutionized parallel computing by offering, for a large and growing set of algorithms, massive data-parallelization on desktop machines. An obstacle to widespread adoption, however, is the difficulty of programming them and the low-level control of the...

Full description

Saved in:
Bibliographic Details
Main Authors: KOLESNICHENKO, Alexey, POSKITT, Christopher M., NANZ, Sebastian, MEYER, Bertrand
Format: text
Language:English
Published: Institutional Knowledge at Singapore Management University 2015
Subjects:
Online Access:https://ink.library.smu.edu.sg/sis_research/4910
https://ink.library.smu.edu.sg/context/sis_research/article/5913/viewcontent/Kolesnichenko_PNM.GPCE.2015__2_.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Singapore Management University
Language: English
id sg-smu-ink.sis_research-5913
record_format dspace
spelling sg-smu-ink.sis_research-59132020-02-13T07:09:32Z Contract-based general-purpose GPU programming KOLESNICHENKO, Alexey POSKITT, Christopher M. NANZ, Sebastian MEYER, Bertrand Using GPUs as general-purpose processors has revolutionized parallel computing by offering, for a large and growing set of algorithms, massive data-parallelization on desktop machines. An obstacle to widespread adoption, however, is the difficulty of programming them and the low-level control of the hardware required to achieve good performance. This paper suggests a programming library, SafeGPU, that aims at striking a balance between programmer productivity and performance, by making GPU data-parallel operations accessible from within a classical object-oriented programming language. The solution is integrated with the design-by-contract approach, which increases confidence in functional program correctness by embedding executable program specifications into the program text. We show that our library leads to modular and maintainable code that is accessible to GPGPU non-experts, while providing performance that is comparable with hand-written CUDA code. Furthermore, runtime contract checking turns out to be feasible, as the contracts can be executed on the GPU. 2015-10-01T07:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/4910 info:doi/10.1145/2814204.2814216 https://ink.library.smu.edu.sg/context/sis_research/article/5913/viewcontent/Kolesnichenko_PNM.GPCE.2015__2_.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University GPGPU parallel computing runtime code generation programming object-orientation design-by-contract program correctness Programming Languages and Compilers Software Engineering
institution Singapore Management University
building SMU Libraries
continent Asia
country Singapore
Singapore
content_provider SMU Libraries
collection InK@SMU
language English
topic GPGPU
parallel computing
runtime code generation
programming
object-orientation
design-by-contract
program correctness
Programming Languages and Compilers
Software Engineering
spellingShingle GPGPU
parallel computing
runtime code generation
programming
object-orientation
design-by-contract
program correctness
Programming Languages and Compilers
Software Engineering
KOLESNICHENKO, Alexey
POSKITT, Christopher M.
NANZ, Sebastian
MEYER, Bertrand
Contract-based general-purpose GPU programming
description Using GPUs as general-purpose processors has revolutionized parallel computing by offering, for a large and growing set of algorithms, massive data-parallelization on desktop machines. An obstacle to widespread adoption, however, is the difficulty of programming them and the low-level control of the hardware required to achieve good performance. This paper suggests a programming library, SafeGPU, that aims at striking a balance between programmer productivity and performance, by making GPU data-parallel operations accessible from within a classical object-oriented programming language. The solution is integrated with the design-by-contract approach, which increases confidence in functional program correctness by embedding executable program specifications into the program text. We show that our library leads to modular and maintainable code that is accessible to GPGPU non-experts, while providing performance that is comparable with hand-written CUDA code. Furthermore, runtime contract checking turns out to be feasible, as the contracts can be executed on the GPU.
format text
author KOLESNICHENKO, Alexey
POSKITT, Christopher M.
NANZ, Sebastian
MEYER, Bertrand
author_facet KOLESNICHENKO, Alexey
POSKITT, Christopher M.
NANZ, Sebastian
MEYER, Bertrand
author_sort KOLESNICHENKO, Alexey
title Contract-based general-purpose GPU programming
title_short Contract-based general-purpose GPU programming
title_full Contract-based general-purpose GPU programming
title_fullStr Contract-based general-purpose GPU programming
title_full_unstemmed Contract-based general-purpose GPU programming
title_sort contract-based general-purpose gpu programming
publisher Institutional Knowledge at Singapore Management University
publishDate 2015
url https://ink.library.smu.edu.sg/sis_research/4910
https://ink.library.smu.edu.sg/context/sis_research/article/5913/viewcontent/Kolesnichenko_PNM.GPCE.2015__2_.pdf
_version_ 1770575093028618240