This HTML5 document contains 45 embedded RDF statements represented using HTML+Microdata notation.

The embedded RDF content will be recognized by any processor of HTML5 Microdata.

Namespace Prefixes

PrefixIRI
dctermshttp://purl.org/dc/terms/
n6http://localhost/temp/predkladatel/
n16http://linked.opendata.cz/resource/domain/vavai/riv/tvurce/
n15http://linked.opendata.cz/resource/domain/vavai/subjekt/
n13http://linked.opendata.cz/ontology/domain/vavai/
n12http://linked.opendata.cz/resource/domain/vavai/vysledek/RIV%2F00216224%3A14330%2F11%3A00054287%21RIV12-MSM-14330___/
n10http://linked.opendata.cz/resource/domain/vavai/zamer/
shttp://schema.org/
skoshttp://www.w3.org/2004/02/skos/core#
n3http://linked.opendata.cz/ontology/domain/vavai/riv/
n2http://linked.opendata.cz/resource/domain/vavai/vysledek/
rdfhttp://www.w3.org/1999/02/22-rdf-syntax-ns#
n7http://linked.opendata.cz/ontology/domain/vavai/riv/klicoveSlovo/
n11http://linked.opendata.cz/ontology/domain/vavai/riv/duvernostUdaju/
xsdhhttp://www.w3.org/2001/XMLSchema#
n14http://linked.opendata.cz/ontology/domain/vavai/riv/jazykVysledku/
n4http://linked.opendata.cz/ontology/domain/vavai/riv/aktivita/
n19http://linked.opendata.cz/ontology/domain/vavai/riv/druhVysledku/
n17http://linked.opendata.cz/ontology/domain/vavai/riv/obor/
n9http://reference.data.gov.uk/id/gregorian-year/

Statements

Subject Item
n2:RIV%2F00216224%3A14330%2F11%3A00054287%21RIV12-MSM-14330___
rdf:type
skos:Concept n13:Vysledek
dcterms:description
When implementing a function mapping on the contemporary GPU, several contradictory performance factors affecting distribution of computation into GPU kernels have to be balanced. A decomposition-fusion scheme suggests to decompose the computational problem to be sol\-ved by several simple functions implemented as standalone kernels and to fuse some of these functions later into more complex kernels to improve memory locality. In this paper, a prototype of source-to-source compiler automating the fusion phase is presented and the impact of fusions generated by the compiler as well as compiler efficiency is experimentally evaluated. When implementing a function mapping on the contemporary GPU, several contradictory performance factors affecting distribution of computation into GPU kernels have to be balanced. A decomposition-fusion scheme suggests to decompose the computational problem to be sol\-ved by several simple functions implemented as standalone kernels and to fuse some of these functions later into more complex kernels to improve memory locality. In this paper, a prototype of source-to-source compiler automating the fusion phase is presented and the impact of fusions generated by the compiler as well as compiler efficiency is experimentally evaluated.
dcterms:title
Automatic Fusions of CUDA-GPU Kernels for Parallel Map Automatic Fusions of CUDA-GPU Kernels for Parallel Map
skos:prefLabel
Automatic Fusions of CUDA-GPU Kernels for Parallel Map Automatic Fusions of CUDA-GPU Kernels for Parallel Map
skos:notation
RIV/00216224:14330/11:00054287!RIV12-MSM-14330___
n13:predkladatel
n15:orjk%3A14330
n3:aktivita
n4:S n4:Z
n3:aktivity
S, Z(MSM0021622419)
n3:cisloPeriodika
4
n3:dodaniDat
n9:2012
n3:domaciTvurceVysledku
n16:8106746 n16:4842138 n16:8419914
n3:druhVysledku
n19:J
n3:duvernostUdaju
n11:S
n3:entitaPredkladatele
n12:predkladatel
n3:idSjednocenehoVysledku
187570
n3:idVysledku
RIV/00216224:14330/11:00054287
n3:jazykVysledku
n14:eng
n3:klicovaSlova
GPU; CUDA; kernels fusion; map; mapped function
n3:klicoveSlovo
n7:CUDA n7:kernels%20fusion n7:mapped%20function n7:GPU n7:map
n3:kodStatuVydavatele
US - Spojené státy americké
n3:kontrolniKodProRIV
[13302F11E310]
n3:nazevZdroje
ACM SIGARCH Computer Architecture News
n3:obor
n17:IN
n3:pocetDomacichTvurcuVysledku
3
n3:pocetTvurcuVysledku
3
n3:rokUplatneniVysledku
n9:2011
n3:svazekPeriodika
39
n3:tvurceVysledku
Fousek, Jan Filipovič, Jiří Madzin, Matúš
n3:zamer
n10:MSM0021622419
s:issn
0163-5964
s:numberOfPages
2
n6:organizacniJednotka
14330