Fabregat-Traver, Diego; Davidović, Davor; Höhnerbach, Markus; Di Napoli, Edoardo (2017) Hybrid CPU-GPU generation of the Hamiltonian and overlap matrices in FLAPW methods. In: Di Napoli, E.; Hermanns, Marc-A.; Iliev, H.; Lintermann, A.; Peyser, A., (eds.) LNCS - High Performance Scientific Computing. pp. 200-211 .
This is the latest version of this item.
|
PDF
- Accepted Version
- article
Download (323kB) | Preview |
Abstract
In this paper we focus on the integration of high-performance numerical libraries in ab initio codes and the portability of performance and scalability. The target of our work is FLEUR, a software for electronic structure calculations developed in the Forschungszentrum J\"ulich over the course of two decades. The presented work follows up on a previous effort to modernize legacy code by re-engineering and rewriting it in terms of highly optimized libraries. We illustrate how this initial effort to get efficient and portable shared-memory code enables fast porting of the code to emerging heterogeneous architectures. More specifically, we port the code to nodes equipped with multiple GPUs. We divide our study in two parts. First, we show considerable speedups attained by minor and relatively straightforward code changes to off-load parts of the computation to the GPUs. Then, we identify further possible improvements to achieve even higher performance and scalability. On a system consisting of 16-cores and 2 GPUs, we observe speedups of up to 5x with respect to our optimized shared-memory code, which in turn means between 7.5x and 12.5x speedup with respect to the original FLEUR code.
| Item Type: | Conference or workshop item published in conference proceedings (UNSPECIFIED) | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Additional Information: | This work was partially funded by the Ministry of Science and Education of the Republic of Croatia and the Deutsche Akademische Austauschdienst (DAAD) from funds of the Bundesministeriums für Bildung und Forschung (BMBF) through project “PPP Kroatien” ID 57216700. Financial support from the Jülich Aachen Research Alliance-High Performance Computing and the Deutsche Forschungsgemeinschaft (DFG) through grant GSC 111 is also gratefully acknowledged. Furthermore, the authors thank the RWTH IT Center and the Jülich Supercomputing Centre for the computational resources. | ||||||||||||||||
| Uncontrolled Keywords: | hybrid computing; GPU; FLEUR; electronic structure | ||||||||||||||||
| Subjects: | NATURAL SCIENCES > Physics > Atomic and Molecular Physics TECHNICAL SCIENCES > Computing TECHNICAL SCIENCES > Computing > Process Computing |
||||||||||||||||
| Divisions: | Center for Informatics and Computing | ||||||||||||||||
| Projects: |
|
||||||||||||||||
| Depositing User: | Davor Davidović | ||||||||||||||||
| Date Deposited: | 15 Sep 2017 14:44 | ||||||||||||||||
| URI: | http://fulir.irb.hr/id/eprint/3575 | ||||||||||||||||
| DOI: | 10.1007/978-3-319-53862-4 |
Available Versions of this Item
-
Hybrid CPU-GPU generation of the Hamiltonian and overlap matrices in FLAPW methods (deposited 21 Mar 2017 11:51)
- Hybrid CPU-GPU generation of the Hamiltonian and overlap matrices in FLAPW methods (deposited 15 Sep 2017 14:44) [Currently Displayed]
Actions (login required)
 |
View Item |
Altmetric
Altmetric
