Eigensolver performance comparison on Cray XC systems
Abstract
Summary Hermitian (symmetric) eigenvalue solvers are the core constituents of electronic structure, quantum‐chemistry, and other HPC applications such as Quantum ESPRESSO, VASP, CP2K, and NWChem to name a few. Our understanding of the performance of symmetric eigenvalue algorithms on various hardware is clearly important to the quantum chemistry or condensed matter physics community but in fact goes beyond that community. For instance, big data analytics is increasingly utilizing eigenvalues solvers, in the study of randomized singular value decomposition (SVD) or principal component analysis (PCA). Noise, vibration, and harshness (NVH) is another field where fast and efficient eigenvalue solvers are required. Most eigenvalue solver packages feature numerous different parameters which can be tuned for performance, eg, the number of nodes, number of total ranks, the decomposition of the matrix, etc. In this paper, we investigate the performance of different packages as well as the influence of these knobs on the solver performance.
- Authors:
-
- National Energy Research Scientific Computing Center Lawrence Berkeley National Laboratory California
- Cray Programming Environments Performance Engineering Cray Inc Minnesota
- Publication Date:
- Sponsoring Org.:
- USDOE
- OSTI Identifier:
- 1473742
- Grant/Contract Number:
- AC02-05CH11231
- Resource Type:
- Publisher's Accepted Manuscript
- Journal Name:
- Concurrency and Computation. Practice and Experience
- Additional Journal Information:
- Journal Name: Concurrency and Computation. Practice and Experience Journal Volume: 31 Journal Issue: 16; Journal ID: ISSN 1532-0626
- Publisher:
- Wiley Blackwell (John Wiley & Sons)
- Country of Publication:
- United Kingdom
- Language:
- English
Citation Formats
Cook, Brandon, Kurth, Thorsten, Deslippe, Jack, Carrier, Pierre, Hill, Nick, and Wichmann, Nathan. Eigensolver performance comparison on Cray XC systems. United Kingdom: N. p., 2018.
Web. doi:10.1002/cpe.4997.
Cook, Brandon, Kurth, Thorsten, Deslippe, Jack, Carrier, Pierre, Hill, Nick, & Wichmann, Nathan. Eigensolver performance comparison on Cray XC systems. United Kingdom. https://doi.org/10.1002/cpe.4997
Cook, Brandon, Kurth, Thorsten, Deslippe, Jack, Carrier, Pierre, Hill, Nick, and Wichmann, Nathan. Tue .
"Eigensolver performance comparison on Cray XC systems". United Kingdom. https://doi.org/10.1002/cpe.4997.
@article{osti_1473742,
title = {Eigensolver performance comparison on Cray XC systems},
author = {Cook, Brandon and Kurth, Thorsten and Deslippe, Jack and Carrier, Pierre and Hill, Nick and Wichmann, Nathan},
abstractNote = {Summary Hermitian (symmetric) eigenvalue solvers are the core constituents of electronic structure, quantum‐chemistry, and other HPC applications such as Quantum ESPRESSO, VASP, CP2K, and NWChem to name a few. Our understanding of the performance of symmetric eigenvalue algorithms on various hardware is clearly important to the quantum chemistry or condensed matter physics community but in fact goes beyond that community. For instance, big data analytics is increasingly utilizing eigenvalues solvers, in the study of randomized singular value decomposition (SVD) or principal component analysis (PCA). Noise, vibration, and harshness (NVH) is another field where fast and efficient eigenvalue solvers are required. Most eigenvalue solver packages feature numerous different parameters which can be tuned for performance, eg, the number of nodes, number of total ranks, the decomposition of the matrix, etc. In this paper, we investigate the performance of different packages as well as the influence of these knobs on the solver performance.},
doi = {10.1002/cpe.4997},
journal = {Concurrency and Computation. Practice and Experience},
number = 16,
volume = 31,
place = {United Kingdom},
year = {Tue Sep 25 00:00:00 EDT 2018},
month = {Tue Sep 25 00:00:00 EDT 2018}
}
https://doi.org/10.1002/cpe.4997
Web of Science
Works referenced in this record:
BerkeleyGW: A massively parallel computer package for the calculation of the quasiparticle and optical properties of materials and nanostructures
journal, June 2012
- Deslippe, Jack; Samsonidze, Georgy; Strubbe, David A.
- Computer Physics Communications, Vol. 183, Issue 6
The ELPA library: scalable parallel eigenvalue solutions for electronic structure theory and computational science
journal, May 2014
- Marek, A.; Blum, V.; Johanni, R.
- Journal of Physics: Condensed Matter, Vol. 26, Issue 21
ScaLAPACK Users' Guide
book, January 1997
- Blackford, L. S.; Choi, J.; Cleary, A.
- Society for Industrial and Applied Mathematics
ABINIT: First-principles approach to material and nanosystem properties
journal, December 2009
- Gonze, X.; Amadon, B.; Anglade, P. -M.
- Computer Physics Communications, Vol. 180, Issue 12
yambo: An ab initio tool for excited state calculations
journal, August 2009
- Marini, Andrea; Hogan, Conor; Grüning, Myrta
- Computer Physics Communications, Vol. 180, Issue 8
Parallel solution of partial symmetric eigenvalue problems from electronic structure calculations
journal, December 2011
- Auckenthaler, T.; Blum, V.; Bungartz, H. -J.
- Parallel Computing, Vol. 37, Issue 12
A Parallel Divide and Conquer Algorithm for the Symmetric Eigenvalue Problem on Distributed Memory Architectures
journal, January 1999
- Tisseur, Françoise; Dongarra, Jack
- SIAM Journal on Scientific Computing, Vol. 20, Issue 6