arrow
Volume 22, Issue 4
GPU-Accelerated LOBPCG Method with Inexact Null-Space Filtering for Solving Generalized Eigenvalue Problems in Computational Electromagnetics Analysis with Higher-Order FEM

A. Dziekonski, M. Rewienski, P. Sypek, A. Lamecki & M. Mrozowski

Commun. Comput. Phys., 22 (2017), pp. 997-1014.

Published online: 2017-10

Export citation
  • Abstract

This paper presents a GPU-accelerated implementation of the Locally Optimal Block Preconditioned Conjugate Gradient (LOBPCG) method with an inexact null-space filtering approach to find eigenvalues in electromagnetics analysis with higher-order FEM. The performance of the proposed approach is verified using the Kepler (Tesla K40c) graphics accelerator, and is compared to the performance of the implementation based on functions from the Intel MKL on the Intel Xeon (E5-2680 v3, 12 threads) central processing unit (CPU) executed in parallel mode. Compared to the CPU reference implementation based on the Intel MKL functions, the proposed GPU-based LOBPCG method with inexact null-space filtering allowed us to achieve up to 2.9-fold acceleration.

  • Keywords

  • AMS Subject Headings

  • Copyright

COPYRIGHT: © Global Science Press

  • Email address
  • BibTex
  • RIS
  • TXT
@Article{CiCP-22-997, author = {}, title = {GPU-Accelerated LOBPCG Method with Inexact Null-Space Filtering for Solving Generalized Eigenvalue Problems in Computational Electromagnetics Analysis with Higher-Order FEM}, journal = {Communications in Computational Physics}, year = {2017}, volume = {22}, number = {4}, pages = {997--1014}, abstract = {

This paper presents a GPU-accelerated implementation of the Locally Optimal Block Preconditioned Conjugate Gradient (LOBPCG) method with an inexact null-space filtering approach to find eigenvalues in electromagnetics analysis with higher-order FEM. The performance of the proposed approach is verified using the Kepler (Tesla K40c) graphics accelerator, and is compared to the performance of the implementation based on functions from the Intel MKL on the Intel Xeon (E5-2680 v3, 12 threads) central processing unit (CPU) executed in parallel mode. Compared to the CPU reference implementation based on the Intel MKL functions, the proposed GPU-based LOBPCG method with inexact null-space filtering allowed us to achieve up to 2.9-fold acceleration.

}, issn = {1991-7120}, doi = {https://doi.org/10.4208/cicp.OA-2016-0155}, url = {http://global-sci.org/intro/article_detail/cicp/9990.html} }
TY - JOUR T1 - GPU-Accelerated LOBPCG Method with Inexact Null-Space Filtering for Solving Generalized Eigenvalue Problems in Computational Electromagnetics Analysis with Higher-Order FEM JO - Communications in Computational Physics VL - 4 SP - 997 EP - 1014 PY - 2017 DA - 2017/10 SN - 22 DO - http://doi.org/10.4208/cicp.OA-2016-0155 UR - https://global-sci.org/intro/article_detail/cicp/9990.html KW - AB -

This paper presents a GPU-accelerated implementation of the Locally Optimal Block Preconditioned Conjugate Gradient (LOBPCG) method with an inexact null-space filtering approach to find eigenvalues in electromagnetics analysis with higher-order FEM. The performance of the proposed approach is verified using the Kepler (Tesla K40c) graphics accelerator, and is compared to the performance of the implementation based on functions from the Intel MKL on the Intel Xeon (E5-2680 v3, 12 threads) central processing unit (CPU) executed in parallel mode. Compared to the CPU reference implementation based on the Intel MKL functions, the proposed GPU-based LOBPCG method with inexact null-space filtering allowed us to achieve up to 2.9-fold acceleration.

A. Dziekonski, M. Rewienski, P. Sypek, A. Lamecki & M. Mrozowski. (2020). GPU-Accelerated LOBPCG Method with Inexact Null-Space Filtering for Solving Generalized Eigenvalue Problems in Computational Electromagnetics Analysis with Higher-Order FEM. Communications in Computational Physics. 22 (4). 997-1014. doi:10.4208/cicp.OA-2016-0155
Copy to clipboard
The citation has been copied to your clipboard