HOMMEXX 1.0: a performance-portable atmospheric dynamical core for the Energy Exascale Earth System Model

Bertagna, Luca; Deakin, Michael; Guba, Oksana; Sunderland, Daniel; Bradley, Andrew M.; Tezaur, Irina K.; Taylor, Mark A.; Salinger, Andrew G.

doi:https://doi.org/10.5194/gmd-12-1423-2019

Articles | Volume 12, issue 4

https://doi.org/10.5194/gmd-12-1423-2019

© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/gmd-12-1423-2019

© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 12, issue 4

Development and technical paper

|

11 Apr 2019

Development and technical paper |

| 11 Apr 2019

HOMMEXX 1.0: a performance-portable atmospheric dynamical core for the Energy Exascale Earth System Model

Luca Bertagna, Michael Deakin, Oksana Guba, Daniel Sunderland, Andrew M. Bradley, Irina K. Tezaur, Mark A. Taylor, and Andrew G. Salinger

Download

Final revised paper (published on 11 Apr 2019)
Preprint (discussion started on 26 Oct 2018)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'Review for HOMMEXX', Anonymous Referee #1, 23 Nov 2018
- AC1: 'Response to referee #1', Luca Bertagna, 01 Feb 2019
RC2: 'HOMMEXX comments', Anonymous Referee #2, 29 Nov 2018
- AC2: 'Response to referee #2', Luca Bertagna, 01 Feb 2019

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Luca Bertagna on behalf of the Authors (06 Feb 2019) Author's response Manuscript

ED: Publish as is (19 Mar 2019) by David Ham

AR by Luca Bertagna on behalf of the Authors (28 Mar 2019) Manuscript

Short summary

We use Kokkos, a C++ library for on-node parallelism, to achieve a performance-portable implementation of HOMME, the atmosphere component of the Earth Energy Exascale System Model. The increasing diversity of high-performance computing (HPC) architectures and the demand for higher resolutions create new challenges when writing efficient code. With Kokkos, we obtain a single code base that performs well on current HPC platforms and enables portable performance to future HPC architectures.