Performance Comparison and applications of energy consumption in parallel with OpenMP and parallel for taskloop

• Luan Pereira UNIPAMPA
• Sandro Matheus Marques UNIPAMPA
• Fábio Rossi IFFAR
• Marcelo Caggiani Luizelli UNIPAMPA
• Arthur Lorenzon UNIPAMPA

DOI: https://doi.org/10.5753/eradrs.2020.10757

Abstract

The OpenMP library has policies that have the same purpose, but with different functionality. An example is the #pragma omp policy is and #pragma omp taskloop. We present a run-time analysis and energy consumption of both directives. We show that the use of the policy is parallel for up to 42% faster and consumes 27.21% less energy than the implementation with taskloop.

Keywords: Evaluation, Performance Measurement and Prediction, Languages, Compilers and Tools Parallel and Distributed Computing

References

Bronis R. de Supinski, M. K. (2019). Openmp technical report 8: Version 5.1 preview. Technical report, OpenMP Architecture Review Board.

Lorenzon, A. F. and Beck Filho, A. C. S. (2019). Parallel Computing Hits the Power Wall Principles, Challenges, and a Survey of Solutions. Springer.

Lorenzon, A. F., De Oliveira, C. C., Souza, J. D., and Beck, A. C. S. (2018). Aurora: Seamless optimization of openmp applications. IEEE Transactions on Parallel and Distributed Systems, 30(5):1007–1021.

Podobas, A. and Karlsson, S. (2016). Towards unifying openmp under the task-parallel paradigm. In International Workshop on OpenMP, pages 116–129. Springer.

Rico, A., Barrera, I. S., Joao, J. A., Randall, J., Casas, M., and Moretó, M. (2019). On the benefits of tasking with openmp. In International Workshop on OpenMP, pages 217–230. Springer.

Teruel, X., Klemm, M., Li, K., Martorell, X., Olivier, S. L., and Terboven, C. (2013). A proposal for task-generating loops in openmp. In International Workshop on OpenMP, pages 1–14. Springer.