Scalability and some optimization of the Finite-volumE Sea ice–Ocean Model, Version 2.0 (FESOM2)
<p>A study of the scalability of the Finite-volumE Sea ice–Ocean circulation Model, Version 2.0 (FESOM2), the first mature global model of its kind formulated on unstructured meshes, is presented. This study includes an analysis of the main computational kernels with a special focus on bottlen...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Copernicus Publications
2019-09-01
|
Series: | Geoscientific Model Development |
Online Access: | https://www.geosci-model-dev.net/12/3991/2019/gmd-12-3991-2019.pdf |
_version_ | 1811212447270305792 |
---|---|
author | N. V. Koldunov N. V. Koldunov V. Aizinger V. Aizinger N. Rakowsky P. Scholz D. Sidorenko S. Danilov S. Danilov T. Jung T. Jung |
author_facet | N. V. Koldunov N. V. Koldunov V. Aizinger V. Aizinger N. Rakowsky P. Scholz D. Sidorenko S. Danilov S. Danilov T. Jung T. Jung |
author_sort | N. V. Koldunov |
collection | DOAJ |
description | <p>A study of the scalability of the Finite-volumE Sea ice–Ocean circulation Model, Version 2.0 (FESOM2), the first mature global model of its kind formulated on unstructured meshes, is presented. This study includes an analysis of the main computational kernels with a special focus on bottlenecks in parallel scalability. Several model enhancements improving this scalability for large numbers of processes are described and tested. Model grids at different resolutions are used on four high-performance computing (HPC) systems with differing computational and communication hardware to demonstrate the model's scalability and throughput. Furthermore, strategies for improvements in parallel performance are presented and assessed. We show that, in terms of throughput, FESOM2 is on a par with state-of-the-art structured ocean models and, in a realistic eddy-resolving configuration (1/10<span class="inline-formula"><sup>∘</sup></span> resolution), can achieve about 16 years per day on 14 000 cores. This suggests that unstructured-mesh models are becoming very competitive tools in high-resolution climate modeling. We show that the main bottlenecks of FESOM2 parallel scalability are the two-dimensional components of the model, namely the computations of the external (barotropic) mode and the sea-ice model. It is argued that these bottlenecks are shared with other general ocean circulation models.</p> |
first_indexed | 2024-04-12T05:29:37Z |
format | Article |
id | doaj.art-e8f1e396cc284ebbb6944101c96f42ca |
institution | Directory Open Access Journal |
issn | 1991-959X 1991-9603 |
language | English |
last_indexed | 2024-04-12T05:29:37Z |
publishDate | 2019-09-01 |
publisher | Copernicus Publications |
record_format | Article |
series | Geoscientific Model Development |
spelling | doaj.art-e8f1e396cc284ebbb6944101c96f42ca2022-12-22T03:46:08ZengCopernicus PublicationsGeoscientific Model Development1991-959X1991-96032019-09-01123991401210.5194/gmd-12-3991-2019Scalability and some optimization of the Finite-volumE Sea ice–Ocean Model, Version 2.0 (FESOM2)N. V. Koldunov0N. V. Koldunov1V. Aizinger2V. Aizinger3N. Rakowsky4P. Scholz5D. Sidorenko6S. Danilov7S. Danilov8T. Jung9T. Jung10MARUM – Center for Marine Environmental Sciences, Leobener Str. 8, 28359 Bremen, GermanyAlfred Wegener Institute, Helmholtz Centre for Polar and Marine Research, Am Handelshafen 12, 27570 Bremerhaven, GermanyAlfred Wegener Institute, Helmholtz Centre for Polar and Marine Research, Am Handelshafen 12, 27570 Bremerhaven, GermanyUniversity of Bayreuth, Chair of Scientific Computing, 95447 Bayreuth, GermanyAlfred Wegener Institute, Helmholtz Centre for Polar and Marine Research, Am Handelshafen 12, 27570 Bremerhaven, GermanyAlfred Wegener Institute, Helmholtz Centre for Polar and Marine Research, Am Handelshafen 12, 27570 Bremerhaven, GermanyAlfred Wegener Institute, Helmholtz Centre for Polar and Marine Research, Am Handelshafen 12, 27570 Bremerhaven, GermanyAlfred Wegener Institute, Helmholtz Centre for Polar and Marine Research, Am Handelshafen 12, 27570 Bremerhaven, GermanyJacobs University, Campus Ring 1, 28759 Bremen, GermanyAlfred Wegener Institute, Helmholtz Centre for Polar and Marine Research, Am Handelshafen 12, 27570 Bremerhaven, GermanyUniversity of Bremen, Bremen, Germany<p>A study of the scalability of the Finite-volumE Sea ice–Ocean circulation Model, Version 2.0 (FESOM2), the first mature global model of its kind formulated on unstructured meshes, is presented. This study includes an analysis of the main computational kernels with a special focus on bottlenecks in parallel scalability. Several model enhancements improving this scalability for large numbers of processes are described and tested. Model grids at different resolutions are used on four high-performance computing (HPC) systems with differing computational and communication hardware to demonstrate the model's scalability and throughput. Furthermore, strategies for improvements in parallel performance are presented and assessed. We show that, in terms of throughput, FESOM2 is on a par with state-of-the-art structured ocean models and, in a realistic eddy-resolving configuration (1/10<span class="inline-formula"><sup>∘</sup></span> resolution), can achieve about 16 years per day on 14 000 cores. This suggests that unstructured-mesh models are becoming very competitive tools in high-resolution climate modeling. We show that the main bottlenecks of FESOM2 parallel scalability are the two-dimensional components of the model, namely the computations of the external (barotropic) mode and the sea-ice model. It is argued that these bottlenecks are shared with other general ocean circulation models.</p>https://www.geosci-model-dev.net/12/3991/2019/gmd-12-3991-2019.pdf |
spellingShingle | N. V. Koldunov N. V. Koldunov V. Aizinger V. Aizinger N. Rakowsky P. Scholz D. Sidorenko S. Danilov S. Danilov T. Jung T. Jung Scalability and some optimization of the Finite-volumE Sea ice–Ocean Model, Version 2.0 (FESOM2) Geoscientific Model Development |
title | Scalability and some optimization of the Finite-volumE Sea ice–Ocean Model, Version 2.0 (FESOM2) |
title_full | Scalability and some optimization of the Finite-volumE Sea ice–Ocean Model, Version 2.0 (FESOM2) |
title_fullStr | Scalability and some optimization of the Finite-volumE Sea ice–Ocean Model, Version 2.0 (FESOM2) |
title_full_unstemmed | Scalability and some optimization of the Finite-volumE Sea ice–Ocean Model, Version 2.0 (FESOM2) |
title_short | Scalability and some optimization of the Finite-volumE Sea ice–Ocean Model, Version 2.0 (FESOM2) |
title_sort | scalability and some optimization of the finite volume sea ice ocean model version 2 0 fesom2 |
url | https://www.geosci-model-dev.net/12/3991/2019/gmd-12-3991-2019.pdf |
work_keys_str_mv | AT nvkoldunov scalabilityandsomeoptimizationofthefinitevolumeseaiceoceanmodelversion20fesom2 AT nvkoldunov scalabilityandsomeoptimizationofthefinitevolumeseaiceoceanmodelversion20fesom2 AT vaizinger scalabilityandsomeoptimizationofthefinitevolumeseaiceoceanmodelversion20fesom2 AT vaizinger scalabilityandsomeoptimizationofthefinitevolumeseaiceoceanmodelversion20fesom2 AT nrakowsky scalabilityandsomeoptimizationofthefinitevolumeseaiceoceanmodelversion20fesom2 AT pscholz scalabilityandsomeoptimizationofthefinitevolumeseaiceoceanmodelversion20fesom2 AT dsidorenko scalabilityandsomeoptimizationofthefinitevolumeseaiceoceanmodelversion20fesom2 AT sdanilov scalabilityandsomeoptimizationofthefinitevolumeseaiceoceanmodelversion20fesom2 AT sdanilov scalabilityandsomeoptimizationofthefinitevolumeseaiceoceanmodelversion20fesom2 AT tjung scalabilityandsomeoptimizationofthefinitevolumeseaiceoceanmodelversion20fesom2 AT tjung scalabilityandsomeoptimizationofthefinitevolumeseaiceoceanmodelversion20fesom2 |