Performance Engineering of the Kernel Polynomal Method on Large-Scale CPU-GPU Systems

Beitrag bei einer Tagung
(Konferenzbeitrag)


Details zur Publikation

Autor(en): Kreutzer M, Hager G, Wellein G, Alvermann A, Fehske H, Pieper A
Herausgeber: IEEE
Jahr der Veröffentlichung: 2015
Tagungsband: Proceedings of the 2015 IEEE International Parallel and Distributed Processing Symposium (IPDPS)
Seitenbereich: 417-426
ISSN: 1530-2075
Sprache: Englisch


Abstract


The Kernel Polynomial Method (KPM) is a well-established scheme in quantum physics and quantum chemistry to determine the eigenvalue density and spectral properties of large sparse matrices. In this work we demonstrate the high optimization potential and feasibility of peta-scale heterogeneous CPU-GPU implementations of the KPM. At the node level we show that it is possible to decouple the sparse matrix problem posed by KPM from main memory bandwidth both on CPU and GPU. To alleviate the effects of scattered data access we combine loosely coupled outer iterations with tightly coupled block sparse matrix multiple vector operations, which enables pure data streaming. All optimizations are guided by a performance analysis and modelling process that indicates how the computational bottlenecks change with each optimization step. Finally we use the optimized node-level KPM with a hybrid-parallel framework to perform large scale heterogeneous electronic structure calculations for novel topological materials on a petascale-class Cray XC30 system.



FAU-Autoren / FAU-Herausgeber

Hager, Georg Dr.
Regionales Rechenzentrum Erlangen (RRZE)
Kreutzer, Moritz
Regionales Rechenzentrum Erlangen (RRZE)
Pieper, Andreas
Regionales Rechenzentrum Erlangen (RRZE)
Wellein, Gerhard Prof. Dr.
Professur für Höchstleistungsrechnen


Autor(en) der externen Einrichtung(en)
Universität Greifswald


Zitierweisen

APA:
Kreutzer, M., Hager, G., Wellein, G., Alvermann, A., Fehske, H., & Pieper, A. (2015). Performance Engineering of the Kernel Polynomal Method on Large-Scale CPU-GPU Systems. In IEEE (Eds.), Proceedings of the 2015 IEEE International Parallel and Distributed Processing Symposium (IPDPS) (pp. 417-426). Hyderabad, India, IN.

MLA:
Kreutzer, Moritz, et al. "Performance Engineering of the Kernel Polynomal Method on Large-Scale CPU-GPU Systems." Proceedings of the Parallel and Distributed Processing Symposium (IPDPS), 2015 IEEE International, Hyderabad, India Ed. IEEE, 2015. 417-426.

BibTeX: 

Zuletzt aktualisiert 2018-10-08 um 05:55