Programação paralela híbrida para CPU e GPU: uma avaliação do OPENACC frente a OPENMP e CUDA
Autor(a) principal: | |
---|---|
Data de Publicação: | 2014 |
Tipo de documento: | Dissertação |
Idioma: | por |
Título da fonte: | Manancial - Repositório Digital da UFSM |
dARK ID: | ark:/26339/001300000807v |
Texto Completo: | http://repositorio.ufsm.br/handle/1/5441 |
Resumo: | As a consequence of the CPU and GPU's architectures advance, in the last years there was a raise of the number of parallel programming APIs for both devices. While OpenMP is used to make parallel programs for the CPU, CUDA and OpenACC are employed in the parallel processing in the GPU. In the programming for the GPU, CUDA presents a model based on functions that make the source code extensive and prone to errors, in addition to leading to low development productivity. OpenACC emerged aiming to solve these problems and to be an alternative to the utilization of CUDA. Similar to OpenMP, this API has policies that ease the development of parallel applications that run on the GPU only. To further increase performance and take advantage of the parallel aspects of both CPU and GPU, it is possible to develop hybrid algorithms that split the processing on the two devices. In that sense, the main objective of this work is to verify if the advantages that OpenACC introduces are also positively reflected on the hybrid programming using OpenMP, if compared to the OpenMP + CUDA model. A second objective of this work is to identify aspects of the two programming models that could limit the performance or on the applications' development. As a way to accomplish these goals, this work presents the development of three hybrid parallel algorithms that are based on the Rodinia's benchmark algorithms, namely, RNG, Hotspot and SRAD, using the hybrid models OpenMP + CUDA and OpenMP + OpenACC. In these algorithms, the CPU part of the code is programmed using OpenMP, while it's assigned for the CUDA and OpenACC the parallel processing on the GPU. After the execution of the hybrid algorithms, the performance, efficiency and the processing's splitting in each one of the devices were analyzed. It was verified, through the hybrid algorithms' runs, that, in the two proposed programming models it was possible to outperform the performance of a parallel application that runs on a single API and in only one of the devices. In addition to that, in the hybrid algorithms RNG and Hotspot, CUDA's performance was superior to that of OpenACC, while in the SRAD algorithm OpenACC was faster than CUDA. |
id |
UFSM_0eb584ac52b2ded29c94c0b61c7cbab1 |
---|---|
oai_identifier_str |
oai:repositorio.ufsm.br:1/5441 |
network_acronym_str |
UFSM |
network_name_str |
Manancial - Repositório Digital da UFSM |
repository_id_str |
|
spelling |
Programação paralela híbrida para CPU e GPU: uma avaliação do OPENACC frente a OPENMP e CUDAHybrid parallel programming for CPU and GPU: an evaluation of OPENACC as RELATED to OPENMP and CUDACPUGPUOpenMPCUDAOpenACCProgramação paralela híbridaDesempenhoOpenMPCUDAHybrid parallel programmingPerformanceCNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAOAs a consequence of the CPU and GPU's architectures advance, in the last years there was a raise of the number of parallel programming APIs for both devices. While OpenMP is used to make parallel programs for the CPU, CUDA and OpenACC are employed in the parallel processing in the GPU. In the programming for the GPU, CUDA presents a model based on functions that make the source code extensive and prone to errors, in addition to leading to low development productivity. OpenACC emerged aiming to solve these problems and to be an alternative to the utilization of CUDA. Similar to OpenMP, this API has policies that ease the development of parallel applications that run on the GPU only. To further increase performance and take advantage of the parallel aspects of both CPU and GPU, it is possible to develop hybrid algorithms that split the processing on the two devices. In that sense, the main objective of this work is to verify if the advantages that OpenACC introduces are also positively reflected on the hybrid programming using OpenMP, if compared to the OpenMP + CUDA model. A second objective of this work is to identify aspects of the two programming models that could limit the performance or on the applications' development. As a way to accomplish these goals, this work presents the development of three hybrid parallel algorithms that are based on the Rodinia's benchmark algorithms, namely, RNG, Hotspot and SRAD, using the hybrid models OpenMP + CUDA and OpenMP + OpenACC. In these algorithms, the CPU part of the code is programmed using OpenMP, while it's assigned for the CUDA and OpenACC the parallel processing on the GPU. After the execution of the hybrid algorithms, the performance, efficiency and the processing's splitting in each one of the devices were analyzed. It was verified, through the hybrid algorithms' runs, that, in the two proposed programming models it was possible to outperform the performance of a parallel application that runs on a single API and in only one of the devices. In addition to that, in the hybrid algorithms RNG and Hotspot, CUDA's performance was superior to that of OpenACC, while in the SRAD algorithm OpenACC was faster than CUDA.Como consequência do avanço das arquiteturas de CPU e GPU, nos últimos anos houve um aumento no número de APIs de programação paralela para os dois dispositivos. Enquanto que OpenMP é utilizada no processamento paralelo em CPU, CUDA e OpenACC são empregadas no processamento paralelo em GPU. Na programação para GPU, CUDA apresenta um modelo baseado em funções que deixam o código fonte extenso e propenso a erros, além de acarretar uma baixa produtividade no desenvolvimento. Objetivando solucionar esses problemas e sendo uma alternativa à utilização de CUDA surgiu o OpenACC. Semelhante ao OpenMP, essa API disponibiliza diretivas que facilitam o desenvolvimento de aplicações paralelas, porém para execução em GPU. Para aumentar ainda mais o desempenho e tirar proveito da capacidade de paralelismo de CPU e GPU, é possível desenvolver algoritmos híbridos que dividam o processamento nos dois dispositivos. Nesse sentido, este trabalho objetiva verificar se as facilidades que o OpenACC introduz também refletem positivamente na programação híbrida com OpenMP, se comparado ao modelo OpenMP + CUDA. Além disso, o trabalho visa relatar as limitações nos dois modelos de programação híbrida que possam influenciar no desempenho ou no desenvolvimento de aplicações. Como forma de cumprir essas metas, este trabalho apresenta o desenvolvimento de três algoritmos paralelos híbridos baseados nos algoritmos do benchmark Rodinia, a saber, RNG, Hotspot e SRAD, utilizando os modelos híbridos OpenMP + CUDA e OpenMP + OpenACC. Nesses algoritmos é atribuída ao OpenMP a execução paralela em CPU, enquanto que CUDA e OpenACC são responsáveis pelo processamento paralelo em GPU. Após as execuções dos algoritmos híbridos foram analisados o desempenho, a eficiência e a divisão da execução em cada um dos dispositivos. Verificou-se através das execuções dos algoritmos híbridos que nos dois modelos de programação propostos foi possível superar o desempenho de uma aplicação paralela em uma única API, com execução em apenas um dos dispositivos. Além disso, nos algoritmos híbridos RNG e Hotspot o desempenho de CUDA foi superior ao desempenho de OpenACC, enquanto que no algoritmo SRAD a API OpenACC apresentou uma execução mais rápida, se comparada à API CUDA.Universidade Federal de Santa MariaBRCiência da ComputaçãoUFSMPrograma de Pós-Graduação em InformáticaStein, Benhur de Oliveirahttp://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4728084T8Charão, Andréa Schwertnerhttp://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4721144D9Cera, Marcia Cristinahttp://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4762397U0Sulzbach, Maurício2015-03-242015-03-242014-08-22info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisapplication/pdfapplication/pdfSULZBACH, Maurício. HYBRID PARALLEL PROGRAMMING FOR CPU AND GPU: AN EVALUATION OF OPENACC AS RELATED TO OPENMP AND CUDA. 2014. 100 f. Dissertação (Mestrado em Ciência da Computação) - Universidade Federal de Santa Maria, Santa Maria, 2014.http://repositorio.ufsm.br/handle/1/5441ark:/26339/001300000807vporinfo:eu-repo/semantics/openAccessreponame:Manancial - Repositório Digital da UFSMinstname:Universidade Federal de Santa Maria (UFSM)instacron:UFSM2022-03-16T15:46:24Zoai:repositorio.ufsm.br:1/5441Biblioteca Digital de Teses e Dissertaçõeshttps://repositorio.ufsm.br/ONGhttps://repositorio.ufsm.br/oai/requestatendimento.sib@ufsm.br||tedebc@gmail.comopendoar:2024-07-29T10:28:52.854636Manancial - Repositório Digital da UFSM - Universidade Federal de Santa Maria (UFSM)false |
dc.title.none.fl_str_mv |
Programação paralela híbrida para CPU e GPU: uma avaliação do OPENACC frente a OPENMP e CUDA Hybrid parallel programming for CPU and GPU: an evaluation of OPENACC as RELATED to OPENMP and CUDA |
title |
Programação paralela híbrida para CPU e GPU: uma avaliação do OPENACC frente a OPENMP e CUDA |
spellingShingle |
Programação paralela híbrida para CPU e GPU: uma avaliação do OPENACC frente a OPENMP e CUDA Sulzbach, Maurício CPU GPU OpenMP CUDA OpenACC Programação paralela híbrida Desempenho OpenMP CUDA Hybrid parallel programming Performance CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO |
title_short |
Programação paralela híbrida para CPU e GPU: uma avaliação do OPENACC frente a OPENMP e CUDA |
title_full |
Programação paralela híbrida para CPU e GPU: uma avaliação do OPENACC frente a OPENMP e CUDA |
title_fullStr |
Programação paralela híbrida para CPU e GPU: uma avaliação do OPENACC frente a OPENMP e CUDA |
title_full_unstemmed |
Programação paralela híbrida para CPU e GPU: uma avaliação do OPENACC frente a OPENMP e CUDA |
title_sort |
Programação paralela híbrida para CPU e GPU: uma avaliação do OPENACC frente a OPENMP e CUDA |
author |
Sulzbach, Maurício |
author_facet |
Sulzbach, Maurício |
author_role |
author |
dc.contributor.none.fl_str_mv |
Stein, Benhur de Oliveira http://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4728084T8 Charão, Andréa Schwertner http://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4721144D9 Cera, Marcia Cristina http://buscatextual.cnpq.br/buscatextual/visualizacv.do?id=K4762397U0 |
dc.contributor.author.fl_str_mv |
Sulzbach, Maurício |
dc.subject.por.fl_str_mv |
CPU GPU OpenMP CUDA OpenACC Programação paralela híbrida Desempenho OpenMP CUDA Hybrid parallel programming Performance CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO |
topic |
CPU GPU OpenMP CUDA OpenACC Programação paralela híbrida Desempenho OpenMP CUDA Hybrid parallel programming Performance CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO |
description |
As a consequence of the CPU and GPU's architectures advance, in the last years there was a raise of the number of parallel programming APIs for both devices. While OpenMP is used to make parallel programs for the CPU, CUDA and OpenACC are employed in the parallel processing in the GPU. In the programming for the GPU, CUDA presents a model based on functions that make the source code extensive and prone to errors, in addition to leading to low development productivity. OpenACC emerged aiming to solve these problems and to be an alternative to the utilization of CUDA. Similar to OpenMP, this API has policies that ease the development of parallel applications that run on the GPU only. To further increase performance and take advantage of the parallel aspects of both CPU and GPU, it is possible to develop hybrid algorithms that split the processing on the two devices. In that sense, the main objective of this work is to verify if the advantages that OpenACC introduces are also positively reflected on the hybrid programming using OpenMP, if compared to the OpenMP + CUDA model. A second objective of this work is to identify aspects of the two programming models that could limit the performance or on the applications' development. As a way to accomplish these goals, this work presents the development of three hybrid parallel algorithms that are based on the Rodinia's benchmark algorithms, namely, RNG, Hotspot and SRAD, using the hybrid models OpenMP + CUDA and OpenMP + OpenACC. In these algorithms, the CPU part of the code is programmed using OpenMP, while it's assigned for the CUDA and OpenACC the parallel processing on the GPU. After the execution of the hybrid algorithms, the performance, efficiency and the processing's splitting in each one of the devices were analyzed. It was verified, through the hybrid algorithms' runs, that, in the two proposed programming models it was possible to outperform the performance of a parallel application that runs on a single API and in only one of the devices. In addition to that, in the hybrid algorithms RNG and Hotspot, CUDA's performance was superior to that of OpenACC, while in the SRAD algorithm OpenACC was faster than CUDA. |
publishDate |
2014 |
dc.date.none.fl_str_mv |
2014-08-22 2015-03-24 2015-03-24 |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/masterThesis |
format |
masterThesis |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
SULZBACH, Maurício. HYBRID PARALLEL PROGRAMMING FOR CPU AND GPU: AN EVALUATION OF OPENACC AS RELATED TO OPENMP AND CUDA. 2014. 100 f. Dissertação (Mestrado em Ciência da Computação) - Universidade Federal de Santa Maria, Santa Maria, 2014. http://repositorio.ufsm.br/handle/1/5441 |
dc.identifier.dark.fl_str_mv |
ark:/26339/001300000807v |
identifier_str_mv |
SULZBACH, Maurício. HYBRID PARALLEL PROGRAMMING FOR CPU AND GPU: AN EVALUATION OF OPENACC AS RELATED TO OPENMP AND CUDA. 2014. 100 f. Dissertação (Mestrado em Ciência da Computação) - Universidade Federal de Santa Maria, Santa Maria, 2014. ark:/26339/001300000807v |
url |
http://repositorio.ufsm.br/handle/1/5441 |
dc.language.iso.fl_str_mv |
por |
language |
por |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf application/pdf |
dc.publisher.none.fl_str_mv |
Universidade Federal de Santa Maria BR Ciência da Computação UFSM Programa de Pós-Graduação em Informática |
publisher.none.fl_str_mv |
Universidade Federal de Santa Maria BR Ciência da Computação UFSM Programa de Pós-Graduação em Informática |
dc.source.none.fl_str_mv |
reponame:Manancial - Repositório Digital da UFSM instname:Universidade Federal de Santa Maria (UFSM) instacron:UFSM |
instname_str |
Universidade Federal de Santa Maria (UFSM) |
instacron_str |
UFSM |
institution |
UFSM |
reponame_str |
Manancial - Repositório Digital da UFSM |
collection |
Manancial - Repositório Digital da UFSM |
repository.name.fl_str_mv |
Manancial - Repositório Digital da UFSM - Universidade Federal de Santa Maria (UFSM) |
repository.mail.fl_str_mv |
atendimento.sib@ufsm.br||tedebc@gmail.com |
_version_ |
1814439749741445120 |