Exploring Opportunities for Performance Improvement in a Global Climate Model

  • Rhuan Costa National Institute for Space Research
  • Celso Luiz Mendes INPE

Resumo


The Brazilian Earth System Model (BESM) is a Global Climate Model (GCM) developed by the Brazilian National Institute for Space Research (INPE). The main purpose of a GCM is to simulate Earth’s climate in a decadal or centennial scale. The simulations usually include representations of the main elements of the Earth, such as atmosphere, ocean, ice and land. Since its first release, BESM has provided support materials for contributions to the Intergovernmental Panel on Climate Change (IPCC). This paper evaluates BESM’s performance and explores optimization possibilities, aiming to speed up the model execution. Our study started with a detailed analysis that characterized the performance of BESM executions on hundreds of processors, which served to reveal the major performance bottlenecks. Next, we worked on schemes to mitigate some of those bottlenecks. The changes made so far resulted on performance gains up to a factor of 4 in some cases, when compared to the way it was previously being executed in production. We also describe ongoing work towards additional performance improvements. Despite presenting results only for BESM, our optimization techniques are applicable to other scientific, multi-physics models as well.

Referências

Casagrande, F., Nobre, P., and Souza, R. B. et al. (2016). Arctic sea ice: Decadal simulations and future scenarios using BESM-OA. Atmospheric and Climate Sciences, 06(02):351–366.

Chou, S. C., Lyra, A., and Mourão, C et al. (2014). Evaluation of the eta simulations nested in three global climate models. American Journal of Climate Change, 03(05):438–454.

Emori, S., Taylor, K., Hewitson, B., Zermoglio, F., Juckes, M., Lautenschlager, M., and Stockhause, M. (2016). Cmip5 data provided at the ipcc data distribution centre.

Griffies, S. and Zadeh, N. (2019). The mom user guide. https://github.com/mom-ocean/MOM5/blob/master/doc/web/user_guide.md

Hammoz (2018) Echam-hammoz - aerosol & atmospheric chemistry modules. https://redmine.hammoz.ethz.ch/projects/hammoz.

Kubota, P. Y. (2015). Visão geral e estado do desenvolvimento mcga-cptec/inpe. In Posters. Workshop BESM 2015: Workshop Anual do Projeto do Modelo Brasileiro do Sistema Terrestre BESM/FAPESP/INCT-MC/Rede CLIMA,.

LNCC (2019) Sdumont - sistema de computação petaflópica do sinapad. http://sdumont.lncc.br/.

Miguel, J. C. H. and Monteiro, M. (2015). Mudanças climáticas, tecnociência e geopolı́tica: Um modelo do sistema terrestre brasileiro e a soberania na produção de futuros climáticos. In Seminários Temáticos, volume 2, page 30. V Reunião de Antropologia da Ciência e da Tecnologia.

Nobre, P., Capistrano, V. B., and Baptista Junior, M. et al. (2016). Modelo brasileiro do sistema terrestre (BESM) para cenários de mudanças climáticas globais. In Modelagem Climática e Vulnerabilidades Setoriais à Mudança do Clima no Brasil, volume 1, pages 33–48. MCTI, Brasilia.

Nobre, P., Siqueira, L. S. P., and Almeida, R. A. F. et al. (2013). Climate simulation and change in the brazilian climate model. Journal of Climate, 26(17):6716–6732.

Rodrigues, E. R., Navaux, P. O. A., Panetta, J., Fazenda, A., Mendes, C. L., and Kale, L. V. (2010). A comparative analysis of load balancing algorithms applied to a weather forecast model. In 2010 22nd International Symposium on Computer Architecture and High Performance Computing, pages 71–78.

Taylor, K. E., Stouffer, R. J., and Meehl, G. A. (2012). An overview of cmip5 and the experiment design. Bulletin of the American Meteorological Society, 93(4):485–498.

Univ. of Wisconsin-Madison (2018) Ibis - integrated biosphere simulator. https://nelson.wisc.edu/sage/data-and-models/lba/ibis.php.
Publicado
08/11/2019
Como Citar

Selecione um Formato
COSTA, Rhuan; MENDES, Celso Luiz. Exploring Opportunities for Performance Improvement in a Global Climate Model. In: SIMPÓSIO EM SISTEMAS COMPUTACIONAIS DE ALTO DESEMPENHO (SSCAD), 20. , 2019, Campo Grande. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2019 . p. 97-108. DOI: https://doi.org/10.5753/wscad.2019.8660.