Optimization and performance evaluation of the IDR iterative Krylov solver on GPUs
Anzt H, Kreutzer M, Ponce E, Peterson GD, Wellein G, Dongarra J (2018)
Publication Status: Published
Publication Type: Journal article
Publication year: 2018
Journal
Publisher: SAGE PUBLICATIONS LTD
Book Volume: 32
Pages Range: 220-230
Journal Issue: 2
Abstract
In this paper, we present an optimized GPU implementation for the induced dimension reduction algorithm. We improve data locality, combine it with an efficient sparse matrix vector kernel, and investigate the potential of overlapping computation with communication as well as the possibility of concurrent kernel execution. A comprehensive performance evaluation is conducted using a suitable performance model. The analysis reveals efficiency of up to 90%, which indicates that the implementation achieves performance close to the theoretically attainable bound.
Authors with CRIS profile
Moritz Kreutzer
Professur für Höchstleistungsrechnen
Gerhard Wellein
Professur für Höchstleistungsrechnen
Related research project(s)
ESSEX - Equipping Sparse Solvers for Exascale
SPP 1648: Software for Exascale Computing
Nov. 1, 2012 - June 30, 2019
Equipping Sparse Solvers for Exascale II (ESSEX-II) (SPPEXA)
SPP 1648: Software for Exascale Computing
Jan. 1, 2016 - Dec. 31, 2018
Involved external institutions
United States (USA) (US)How to cite
APA:
Anzt, H., Kreutzer, M., Ponce, E., Peterson, G.D., Wellein, G., & Dongarra, J. (2018). Optimization and performance evaluation of the IDR iterative Krylov solver on GPUs. International Journal of High Performance Computing Applications, 32(2), 220-230. https://dx.doi.org/10.1177/1094342016646844
MLA:
Anzt, Hartwig, et al. "Optimization and performance evaluation of the IDR iterative Krylov solver on GPUs." International Journal of High Performance Computing Applications 32.2 (2018): 220-230.
BibTeX: Download