Análise de uma Aplicação de Modelagem Atmosférica em Nuvem e em Contêineres Utilizando Rastros

• Lucas de Araujo UFSM
• Andrea Charão UFSM
• João Vicente Lima UFSM
• Haroldo de Campos Velho INPE

DOI: https://doi.org/10.5753/wscad_estendido.2020.14089

Abstract

Exploring technologies such as the cloud and containers to execute high performance parallel applications might be beneficial, mainly to research in this area. Besides executing the applications, it is important to obtain metrics to observe the behavior of the application in each environment. In this context, traces were used to visualize the communication and analyze the performance of an atmospheric modeling application in these environments. The results point out a low overhead to the application executed and that the containers distribution might affect the performance.

References

Eriksson, J., Ojeda-May, P., Ponweiser, T., and Steinreiter, T. (2016). Proling and tracing tools for performance analysis of large scale applications. PRACE: Partnership for Advanced Computing in Europe.

Evangelinos, C. and Hill, C. N. (2008). Cloud computing for parallel scientic HPC applications: Feasibility of running coupled atmosphere-ocean climate models on Amazon's EC2. In In The 1st Workshop on Cloud Computing and its Applications (CCA).

Freitas, S. R., Panetta, J., Longo, K. M., Rodrigues, L. F., Moreira, D. S., Rosário, N. E., Silva Dias, P. L., Silva Dias, M. A. F., Souza, E. P., Freita, E. D., Longo, M., Frassoni, A., Fazenda, A. L., Silva, C. M. S., Pavani, C. A. B., Eiras, D., França, D. A., Massaru, D., Silva, F. B., Santos, F. C., Pereira, G., Camponogara, G., Ferrada, G. A., Campos Velho, H. F., Menezes, I., Freire, J. L. F., Alonso, M., Gacita, M. S., Zarzur, M. Z., Fonseca, R. M., Lima, R. S., Siqueira, R. A., Braz, R., Tomita, S., Oliveira, V., and Martins, L. D. (2017). The brazilian developments on the regional atmospheric modeling system (BRAMS 5.2): an integrated environmental model tuned for tropical areas. Geoscientic Model Develpment, 10:189–222.

He, Q., Zhou, S., Kobler, B., Duffy, D., and McGlynn, T. (2010). Case study for running In Proceedings of the 19th ACM International HPC applications in public clouds. Symposium on High Performance Distributed Computing, HPDC '10, page 395–401, New York, NY, USA. Association for Computing Machinery.

Kurtzer, G. M., Sochat, V., and Bauer, M. W. (2017). Singularity: Scientic containers for mobility of compute. PloS one, 12(5):e0177459.

Merkel, D. (2014). Docker: lightweight linux containers for consistent development and deployment. Linux journal, 2014(239):2.

Rodrigues, E. R., Navaux, P. O., 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. IEEE.

Rodrigues, L. F., Lima, S. T., Ruiz, R., Panetta, J., de Freitas, S. R., and Campos Velho, H. F. (2019). Parallel version for the BRAMS with Runge-Kutta dynamical core. Conference of Computational Interdisciplinary Science.

Saha, P., Beltre, A., Uminski, P., and Govindaraju, M. (2018). Evaluation of docker containers for scientic workloads in the cloud. In Proceedings of the Practice and Experience on Advanced Research Computing, pages 1–8.

Silva Junior, M. B., Panetta, J., and Stephany, S. (2017). Portability with efciency of the advection of BRAMS between multi-core and many-core architectures. Concurrency and Computation: Practice and Experience, 29(22):e3959.

Younge, A. J., Pedretti, K., Grant, R. E., and Brightwell, R. (2017). A tale of two systems: Using containers to deploy HPC applications on supercomputers and clouds. In 2017 IEEE International Conference on Cloud Computing Technology and Science (CloudCom), pages 74–81.