Other formats:
BibTeX
LaTeX
RIS
@inproceedings{986495, author = {Filipovič, Jiří and Fousek, Jan and Lakomý, Bedřich and Madzin, Matúš}, address = {LOS ALAMITOS, CA, USA}, booktitle = {Symposium on Application Accelerators in High Performance Computing}, keywords = {GPGPU; code optimization; kernel fusion; FEM}, howpublished = {elektronická verze "online"}, language = {eng}, location = {LOS ALAMITOS, CA, USA}, isbn = {978-1-4673-2882-1}, pages = {141-144}, publisher = {IEEE}, title = {Automatically Optimized GPU Acceleration of Element Subroutines in Finite Element Method}, year = {2012} }
TY - JOUR ID - 986495 AU - Filipovič, Jiří - Fousek, Jan - Lakomý, Bedřich - Madzin, Matúš PY - 2012 TI - Automatically Optimized GPU Acceleration of Element Subroutines in Finite Element Method PB - IEEE CY - LOS ALAMITOS, CA, USA SN - 9781467328821 KW - GPGPU KW - code optimization KW - kernel fusion KW - FEM N2 - The element subroutines in finite element method (FEM) provides enough parallelism to be successfully accelerated by contemporary GPUs. However, their efficient implementation is not straightforward and requires time-consuming exploration of numerous implementation variants. In this paper, we present optimization by kernel fusion for element subroutines. Moreover, we show how the optimization is automated using our source-to-source compiler. We demonstrate the optimization of the element subroutines for FEM model using St.\,Venant-Kirchhoff material. The performance of code generated by our compiler outperforms our previously published hand-tuned implementation by factor of 1.32 -- 1.54 depending on used GPU architecture. Although the optimization technique is demonstrated on element subroutines for using St.\,Venant-Kirchhoff material, it is generally usable for wider area of computationally-demanding problems. ER -
FILIPOVIČ, Jiří, Jan FOUSEK, Bedřich LAKOMÝ and Matúš MADZIN. Automatically Optimized GPU Acceleration of Element Subroutines in Finite Element Method. Online. In \textit{Symposium on Application Accelerators in High Performance Computing}. LOS ALAMITOS, CA, USA: IEEE, 2012, p.~141-144. ISBN~978-1-4673-2882-1.
|