Improving Performance and Energy Efficiency of the Fletcher Method for Oil Extraction Simulation

• Matheus Serpa UFRGS
• Pablo Pavan UFRGS
• Jairo Panetta ITA
• Alexandre Carissimi UFRGS
• Philippe Olivier Alexandre Navaux UFRGS

DOI: https://doi.org/10.5753/eradrs.2020.10777

Abstract

Fletcher method is the basis of oil extraction simulation tools used by the industry. To perform such simulations, parallel architectures are used, providing results faster and with greater accuracy. However, to achieve high performance on these architectures, several challenges must be taken into consideration. In this paper, we improve the performance and energy efficiency of the Fletcher Method up to 73.6% and 42.6%.

Keywords: Parallel Applications to Problems of Real Solutions, Evaluation, Performance Measurement and Prediction

References

Lukawski, M. Z., Anderson, B. J., Augustine, C., Capuano Jr, L. E., Beckers, K. F., Livesay, B., and Tester, J. W. (2014). Cost Analysis of Oil, Gas, and Geothermal Well Drilling. Journal of Petroleum Science and Engineering, 118:1–14.

Serpa, M. S., Cruz, E. H., Diener, M., Krause, A. M., Navaux, P. O. A., Panetta, J., Farrés, A., Rosas, C., and Hanzich, M. (2019). Optimization strategies for geophysics models on manycore systems. The International Journal of High Performance Computing Applications, 33(3):473–486.