Uma análise do uso de containers para portabilidade de código para GPU na nuvem computacional
Abstract
Recent advances in virtualization technologies are enabling the execution of GPU code on cloud computing services. Moreover, container technologies are facilitating the development and migration of high-performance application to the cloud. In this work, we leverage the Singularity container to facilitate the migration of seismic processing GPU code and analyze its impact on performance. Our results indicate that the Singularity container does not add performance overhead to applications that were parallelized with CUDA, OpenCL and OpenMP.References
Kurtzer, G. M., Sochat, V., and Bauer, M. W. (2017). Singularity: Scientific containers for mobility of compute. PloS one, 12(5):e0177459.
Mann, J., Jäger, R., Müller, T., Höcht, G., and Hubral, P. (1999). Common-reflectionsurface stack—a real data example. Journal of applied geophysics, 42(3-4):301–318.
Pahl, C. (2015). Containerization and the paas cloud. IEEE Cloud Computing, 2(3):24–31.
Mann, J., Jäger, R., Müller, T., Höcht, G., and Hubral, P. (1999). Common-reflectionsurface stack—a real data example. Journal of applied geophysics, 42(3-4):301–318.
Pahl, C. (2015). Containerization and the paas cloud. IEEE Cloud Computing, 2(3):24–31.
Published
2018-04-13
How to Cite
BRUNETTA, Jeferson Rech; RODAMILANS, Charles Boulhosa; BENEDICTO, Caian; BORIN, Edson.
Uma análise do uso de containers para portabilidade de código para GPU na nuvem computacional. In: REGIONAL SCHOOL OF HIGH PERFORMANCE COMPUTING FROM SÃO PAULO (ERAD-SP), 9. , 2018, São José dos Campos.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2018
.
p. 33-36.
DOI: https://doi.org/10.5753/eradsp.2018.13596.
