Avaliação de Desempenho no Supercomputador SDumont de uma Estratégia de Decomposição de Domínio usando as Funcionalidades de Mapeamento Topológico do MPI para um Método Numérico de Escoamento de Fluidos.

  • Stiw Herrera LNCC
  • Weber Ribeiro LNCC
  • Thiago Teixeira LNCC
  • André Carneiro LNCC
  • Frederico Cabral LNCC
  • Márcio Borges LNCC
  • Carla Osthoff LNCC

Resumo


Oil and gas simulations need new high-performance computing techniques to deal with the large amount of data allocation and the high computational cost that we obtain from the numerical method. The domain decomposition technique (domain division technique) was applied to a three-dimensional oil reservoir, where the MPI (Message Passing Interface) allowed the creation of a uni, bi and three-dimensional topology, where a subdivision of a reservoir could be solved in each MPI process created. A performance study was developed with these domain decomposition strategies in 20 computational nodes of the SDumont Supercomputer, using a Cascade Lake architecture.
Palavras-chave: Aplicações em Agricultura, Biologia, Engenharia, Física, Matemática, Medicina, Mercado Financeiro, Nanociências, Óleo e Gás, Química e outras áreas

Referências

Bjørstad, P. E., Brenner, S. C., Halpern, L., Kim, H. H., Kornhuber, R., Rahman, T., and Widlund, O. B. (2018). Domain Decomposition Methods in Science and Engineering XXIV. Springer.

Carneiro, A. R., Bez, J. L., Boito, F. Z., Fagundes, B. A., Osthoff, C., and Navaux, P. O. A. (2018a). Collective i/o performance on the santos dumont supercomputer. In 2018 26th Euromicro International Conference on Parallel, Distributed and Network-based Processing (PDP), pages 45–52. IEEE.

Carneiro, A. R., Bez, J. L., Boito, F. Z., Fagundes, B. A., Osthoff, C., and Navaux, P. O. A. (2018b). Collective i/o performance on the santos dumont supercomputer. In 2018 26th Euromicro International Conference on Parallel, Distributed and Network-based Processing (PDP), pages 45–52. IEEE

Chen, Z. (2007). Reservoir simulation: mathematical techniques in oil recovery. SIAM.

Correa, M. and Borges, M. (2013). A semi-discrete central scheme for scalar hyperbolic conservation laws with heterogeneous storage coefficient and its application to porous media flow. International Journal for Numerical Methods in Fluids, 73(3):205–224.

Lima, I. d. C. M. (2017). Simulação de reservatórios de petróleo em paralelo utilizando malhas não-estruturadas 2d e 3d.

Malyshkin, V. (2017). Parallel Computing Technologies: 14th International Conference, PaCT 2017, Nizhny Novgorod, Russia, September 4-8, 2017, Proceedings, volume 10421. Springer.

Murad, M. A., Borges, M., Obregon, J. A., and Correa, M. (2013). A new locally conservative ´ numerical method for two-phase flow in heterogeneous poroelastic media. Computers and Geotechnics, 48:192–207.

Palin, M. F. (2007). Tecnicas de decomposição de domínio em computaçao paralela para simulação de campos eletromagnéticos pelo método dos elementos finitos. PhD thesis, Universidade de São Paulo

Parashar, M. and Yotov, I. (1998). An environment for parallel multi-block, multi-resolution reservoir simulations. In Proceedings of the 11th International Conference on Parallel and Distributed Computing Systems (PDCS 98), Chicago, IL, International Society for Computers and their Applications (ISCA), pages 230–235.

Straatsma, T. P., Antypas, K. B., and Williams, T. J. (2017). Exascale scientific applications: Scalability and performance portability. CRC Press

Tuane, V. L. (2012). Simulação numérica tridimensional de escoamento em reservátorios de petróleo heterogêneos. Master’s thesis, LNCC/MCT, Petrópolis, RJ, Brasil

Winkelmann, R., Hauser, J., and Williams, R. D. (1999). Strategies for parallel and numerical ¨ scalability of cfd codes. Computer methods in applied mechanics and engineering, 174(3- 4):433–456.
Publicado
30/11/2020
HERRERA, Stiw; RIBEIRO, Weber ; TEIXEIRA, Thiago ; CARNEIRO, André; CABRAL, Frederico; BORGES, Márcio ; OSTHOFF, Carla . Avaliação de Desempenho no Supercomputador SDumont de uma Estratégia de Decomposição de Domínio usando as Funcionalidades de Mapeamento Topológico do MPI para um Método Numérico de Escoamento de Fluidos.. In: ESCOLA REGIONAL DE ALTO DESEMPENHO DO RIO DE JANEIRO (ERAD-RJ), 6. , 2020, Evento Online. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2020 . p. 31-35. DOI: https://doi.org/10.5753/eradrj.2020.14513.