Application of Intel Parallel Studio tools for code modernization for finite difference numerical methods to solve partial differential equations in Haswell / Broadwell architecture

• Gabriel P. da Costa CEFET/RJ
• Carla Osthoff LNCC
• Frederico Cabral LNCC

Abstract

This papers aim to show a numerical method code optimization strategy for Intel parallel environments. Will be described the standard and optimized code versions, as well the reasons that led to the development of this new version and the tests that prove the performance improvement and confirm its validity

Keywords: Partial Differential Equation, Numerical Methods, Code Modernization

References

Cabral, F. L.; Osthoff, C.; Costa, G. P.; Brandão, D.; Kischinhevsky, M.; de Oliveira, S. L. G. (2017). Tuning up tvd hopmoc method on intel mic xeon phi architectures with intel parallel studio tools. In 2017 International Symposium on Computer Architecture and High Performance Computing Workshops (SBAC-PADW), pages 19–24. IEEE.

Cabral, F. L.; Osthoff, C.; Costa, G. P.; de Oliveira, S. L. G.; Brandão, D.; Kischinhevsky, M. (2018a). An openmp implementation of the tvd–hopmoc method based on a synchronization mechanism using locks between adjacent threads on xeon phi (tm) accelerators. In International Conference on Computational Science, pages 701–707. Springer.

Cabral, F. L.; Osthoff, C.; Souto, R. P.; Costa, G. P.; de Oliveira, S. L. G.; Brandão, D.; Kischinhevsky, M. (2018b). Fine-tuning an openmp-based tvd–hopmoc method using intel R parallel studio xe tools on intel xeon architectures. In Latin American High Performance Computing Conference, pages 194–209. Springer.