Sökning: id:"swepub:oai:research.chalmers.se:6a40c7cf-bcc5-466a-9df6-7d9677b4f978" >
Counting and Occurr...
Counting and Occurrence Sort for GPUs using an Embedded Language
-
- Svenningsson, Josef, 1976 (författare)
- Chalmers tekniska högskola,Chalmers University of Technology
-
- Svensson, Joel Bo, 1979 (författare)
- Chalmers tekniska högskola,Chalmers University of Technology
-
- Sheeran, Mary, 1959 (författare)
- Chalmers tekniska högskola,Chalmers University of Technology
-
(creator_code:org_t)
- ISBN 9781450323819
- 2013-09-23
- 2013
- Engelska.
-
Ingår i: The 2nd ACM SIGPLAN Workshop on Functional High-Performance Computing, FHPC'13. - New York, NY, USA : ACM. - 9781450323819 ; 48:12, s. 37-45
- Relaterad länk:
-
http://dx.doi.org/10...
-
visa fler...
-
https://research.cha...
-
https://doi.org/10.1...
-
visa färre...
Abstract
Ämnesord
Stäng
- This paper investigates two sorting algorithms: counting sort and avariation, occurrence sort, which also removes duplicate elements,and examines their suitability for running on the GPU. The dupli-cate removing variation turns out to have a natural functional, data-parallel implementation which makes it particularly interesting forGPUs.The algorithms are implemented in Obsidian, a high-level do-main specific language for GPU programming.Measurements show that our implementations in many casesoutperform the sorting algorithm provided by the library Thrust.Furthermore, occurrence sort is another factor of two faster thanordinary counting sort. We conclude that counting sort is an impor-tant contender when considering sorting algorithms for the GPU,and that occurrence sort is highly preferable when applicable. Wealso show that Obsidian can produce very competitive code.
Ämnesord
- NATURVETENSKAP -- Data- och informationsvetenskap -- Programvaruteknik (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Software Engineering (hsv//eng)
Nyckelord
- Sorting
- Embedded language
Publikations- och innehållstyp
- kon (ämneskategori)
- ref (ämneskategori)
Hitta via bibliotek
Till lärosätets databas