Computational strategies for implementation of 2D elastic wave modeling in GPU
Elastic wave modeling presents a challenge to implement since it is a computationally costly procedure. Nowadays, due to GPU increased power jointly with development in HPC computation, it is possible to execute elastic modeling with better execution times and memory use. This study evaluates the pe...
- Autores:
- Tipo de recurso:
- Fecha de publicación:
- 2020
- Institución:
- Universidad Católica de Pereira
- Repositorio:
- Repositorio Institucional - RIBUC
- Idioma:
- spa
- OAI Identifier:
- oai:repositorio.ucp.edu.co:10785/13684
- Acceso en línea:
- https://revistas.ucp.edu.co/index.php/entrecienciaeingenieria/article/view/2016
http://hdl.handle.net/10785/13684
- Palabra clave:
- Rights
- openAccess
- License
- Derechos de autor 2021 Entre Ciencia e Ingeniería
| id |
RepoRIBUC_0a80996fafbbed85624514d261134196 |
|---|---|
| oai_identifier_str |
oai:repositorio.ucp.edu.co:10785/13684 |
| network_acronym_str |
RepoRIBUC |
| network_name_str |
Repositorio Institucional - RIBUC |
| repository_id_str |
|
| spelling |
Computational strategies for implementation of 2D elastic wave modeling in GPUEstrategias computacionales para la implementación de modelado elástico 2D sobre GPUElastic wave modeling presents a challenge to implement since it is a computationally costly procedure. Nowadays, due to GPU increased power jointly with development in HPC computation, it is possible to execute elastic modeling with better execution times and memory use. This study evaluates the performance of 2 strategies for implementing elastic modeling using different kernel launching layouts, CPML memory allocation strategies, and wavefield storage management. The performance measures show that the algorithm, which includes 2D kernel launching layout, CPML reduced memory strategy, and GPU global memory storage to save wavefield cube peaks up to 88.4% better execution time and uses 13.3 times less memory to obtain the same elastic modeling results. There is also an increasing trend of enhancement in execution times and memory savings when working with models of bigger sizes with this strategy.El modelado de onda elástico presenta un reto de implementación debido a que es un procedimiento computacionalmente costoso. En la actualidad, debido al incremento en la potencia en GPU junto con el desarrollo de la computación HPC, es posible ejecutar modelado elástico con mejores tiempos de ejecución y uso de memoria. Este estudio evalúa el desempeño de 2 estrategias para implementar modelado elástico usando diferentes diseños para ejecución de kernel, estrategias de asignación de memoria para el cálculo de CPML y administración del almacenamiento del campo de onda. Las mediciones de desempeño muestran que el algoritmo que incluye diseño de ejecución de kernel 2D, la estrategia de memoria reducida CPML y el almacenamiento en memoria global de GPU del campo de onda alcanza un máximo de 88.4% mejor tiempo de ejecución y utiliza un 13.3 veces menos memoria para obtener los mismos resultados de modelado elástico. Existe también una creciente tendencia de mejora de tiempo de ejecución y ahorro de memoria cuando se trabaja con modelos de tamaños más grandes con esta estrategia.Universidad Católica de Pereira2020-12-312023-08-29T03:49:39Z2023-08-29T03:49:39ZArtículo de revistahttp://purl.org/coar/resource_type/c_6501http://purl.org/coar/version/c_970fb48d4fbd8a85info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionhttp://purl.org/coar/resource_type/c_2df8fbb1application/pdfhttps://revistas.ucp.edu.co/index.php/entrecienciaeingenieria/article/view/201610.31908/19098367.2016http://hdl.handle.net/10785/13684Entre ciencia e ingeniería; Vol 14 No 28 (2020); 52-58Entre Ciencia e Ingeniería; Vol. 14 Núm. 28 (2020); 52-58Entre ciencia e ingeniería; v. 14 n. 28 (2020); 52-582539-41691909-8367spahttps://revistas.ucp.edu.co/index.php/entrecienciaeingenieria/article/view/2016/1866Derechos de autor 2021 Entre Ciencia e Ingenieríahttps://creativecommons.org/licenses/by-nc/4.0/deed.es_ESinfo:eu-repo/semantics/openAccesshttp://purl.org/coar/access_right/c_abf2Páez Chanagá, AndersonRamirez Silva, Ana BeatrizSánchez Galvis, Ivan Javieroai:repositorio.ucp.edu.co:10785/136842025-01-27T23:30:51Z |
| dc.title.none.fl_str_mv |
Computational strategies for implementation of 2D elastic wave modeling in GPU Estrategias computacionales para la implementación de modelado elástico 2D sobre GPU |
| title |
Computational strategies for implementation of 2D elastic wave modeling in GPU |
| spellingShingle |
Computational strategies for implementation of 2D elastic wave modeling in GPU |
| title_short |
Computational strategies for implementation of 2D elastic wave modeling in GPU |
| title_full |
Computational strategies for implementation of 2D elastic wave modeling in GPU |
| title_fullStr |
Computational strategies for implementation of 2D elastic wave modeling in GPU |
| title_full_unstemmed |
Computational strategies for implementation of 2D elastic wave modeling in GPU |
| title_sort |
Computational strategies for implementation of 2D elastic wave modeling in GPU |
| description |
Elastic wave modeling presents a challenge to implement since it is a computationally costly procedure. Nowadays, due to GPU increased power jointly with development in HPC computation, it is possible to execute elastic modeling with better execution times and memory use. This study evaluates the performance of 2 strategies for implementing elastic modeling using different kernel launching layouts, CPML memory allocation strategies, and wavefield storage management. The performance measures show that the algorithm, which includes 2D kernel launching layout, CPML reduced memory strategy, and GPU global memory storage to save wavefield cube peaks up to 88.4% better execution time and uses 13.3 times less memory to obtain the same elastic modeling results. There is also an increasing trend of enhancement in execution times and memory savings when working with models of bigger sizes with this strategy. |
| publishDate |
2020 |
| dc.date.none.fl_str_mv |
2020-12-31 2023-08-29T03:49:39Z 2023-08-29T03:49:39Z |
| dc.type.none.fl_str_mv |
Artículo de revista http://purl.org/coar/resource_type/c_6501 http://purl.org/coar/version/c_970fb48d4fbd8a85 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion |
| dc.type.coar.fl_str_mv |
http://purl.org/coar/resource_type/c_2df8fbb1 |
| status_str |
publishedVersion |
| dc.identifier.none.fl_str_mv |
https://revistas.ucp.edu.co/index.php/entrecienciaeingenieria/article/view/2016 10.31908/19098367.2016 http://hdl.handle.net/10785/13684 |
| url |
https://revistas.ucp.edu.co/index.php/entrecienciaeingenieria/article/view/2016 http://hdl.handle.net/10785/13684 |
| identifier_str_mv |
10.31908/19098367.2016 |
| dc.language.none.fl_str_mv |
spa |
| language |
spa |
| dc.relation.none.fl_str_mv |
https://revistas.ucp.edu.co/index.php/entrecienciaeingenieria/article/view/2016/1866 |
| dc.rights.none.fl_str_mv |
Derechos de autor 2021 Entre Ciencia e Ingeniería https://creativecommons.org/licenses/by-nc/4.0/deed.es_ES info:eu-repo/semantics/openAccess http://purl.org/coar/access_right/c_abf2 |
| rights_invalid_str_mv |
Derechos de autor 2021 Entre Ciencia e Ingeniería https://creativecommons.org/licenses/by-nc/4.0/deed.es_ES http://purl.org/coar/access_right/c_abf2 |
| eu_rights_str_mv |
openAccess |
| dc.format.none.fl_str_mv |
application/pdf |
| dc.publisher.none.fl_str_mv |
Universidad Católica de Pereira |
| publisher.none.fl_str_mv |
Universidad Católica de Pereira |
| dc.source.none.fl_str_mv |
Entre ciencia e ingeniería; Vol 14 No 28 (2020); 52-58 Entre Ciencia e Ingeniería; Vol. 14 Núm. 28 (2020); 52-58 Entre ciencia e ingeniería; v. 14 n. 28 (2020); 52-58 2539-4169 1909-8367 |
| institution |
Universidad Católica de Pereira |
| repository.name.fl_str_mv |
|
| repository.mail.fl_str_mv |
|
| _version_ |
1844494706672664576 |