Performance comparison between CPU and GPU executions of the 3D CSEM method implementation

  • Rômulo T. Lima LNCC / UCP
  • Mateus F. Lima de Souza LNCC / CEFET-RJ
  • Ramoni Z. S. Azevedo LNCC
  • Antônio Tadeu A. Gomes LNCC
  • Roberto P. Souto LNCC
  • Tiziano Labruzzo LNCC / Zlemlink Ltda
  • Andrea Zerilli LNCC / Zlemlink Ltda

Abstract


A CPU vs. GPU performance comparison was performed for an implementation of the 3D CSEM method, which is a geophysical technique used to locate oil reserves on the seabed. A routine from the PETSc library dominates the total execution time, and its call can be executed on either the CPU or the GPU.

References

Adhianto, L., Banerjee, S., Fagan, M., Krentel, M., Marin, G., Mellor-Crummey, J., and Tallent, N. R. (2010). HPCToolkit: tools for performance analysis of optimized parallel programs. Concurrency and Computation: Practice and Experience, 22(6):685–701.

Balay, S., Gropp, W. D., McInnes, L. C., and Smith, B. F. (1997). Efficient management of parallelism in object oriented numerical software libraries. In Arge, E., Bruaset, A. M., and Langtangen, H. P., editors, Modern Software Tools in Scientific Computing, pages 163–202. Birkhäuser Press.

de Souza, M., Lima, R., Souto, R. P., Gomes, A. T. A., Labruzzo, T., and Zerilli, A. (2023). Avaliação de desempenho de implementação paralela do método CSEM 3D no supercomputador Santos Dumont. In Anais da VIII Escola Regional de Alto Desempenho do Rio de Janeiro, pages 29–31, Porto Alegre, RS, Brasil. SBC.

Zerilli, A., Labruzzo, T., Zanzi, M., Buonora, M. P., Crepaldi, J. L., and Menezes, P. T. (2014). Broadband marine CSEM: New benefits for subsalt and around salt exploration. In 2014 SEG Annual Meeting. OnePetro.
Published
2024-11-05
LIMA, Rômulo T.; SOUZA, Mateus F. Lima de; AZEVEDO, Ramoni Z. S.; GOMES, Antônio Tadeu A.; SOUTO, Roberto P.; LABRUZZO, Tiziano; ZERILLI, Andrea. Performance comparison between CPU and GPU executions of the 3D CSEM method implementation. In: REGIONAL SCHOOL OF HIGH PERFORMANCE COMPUTING FROM RIO DE JANEIRO (ERAD-RJ), 9. , 2024, Rio de Janeiro/RJ. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2024 . p. 28-30. DOI: https://doi.org/10.5753/eradrj.2024.4407.