Reproducibility of an Earth System Model under a Change in Computing Environment

Most Earth System Models (ESMs) are running under different high performance computing (HPC) environments. This has several advantages, from allowing different groups to work with the same tool in parallel, to leveraging the burden of ensemble climate simulations but also offering alternative soluti...

Full description

Bibliographic Details
Main Authors: Massonnet, François, Ménégoz, Martin, Acosta, Mario, Yepes-Arbós, Xavier, Exarchout, Eleftheria, Doblas-Reyes, Francisco
Other Authors: UCL - SST/ELI/ELIC - Earth & Climate
Format: Report
Language:English
Published: 2018
Subjects:
Online Access:http://hdl.handle.net/2078.1/203820
id ftunistlouisbrus:oai:dial.uclouvain.be:boreal:203820
record_format openpolar
spelling ftunistlouisbrus:oai:dial.uclouvain.be:boreal:203820 2024-05-12T07:55:47+00:00 Reproducibility of an Earth System Model under a Change in Computing Environment Massonnet, François Ménégoz, Martin Acosta, Mario Yepes-Arbós, Xavier Exarchout, Eleftheria Doblas-Reyes, Francisco UCL - SST/ELI/ELIC - Earth & Climate 2018 http://hdl.handle.net/2078.1/203820 eng eng boreal:203820 BSC Technical Memorandum http://hdl.handle.net/2078.1/203820 info:eu-repo/semantics/openAccess High Performance Computing Climate model Reproducibility info:eu-repo/semantics/report 2018 ftunistlouisbrus 2024-04-18T17:30:21Z Most Earth System Models (ESMs) are running under different high performance computing (HPC) environments. This has several advantages, from allowing different groups to work with the same tool in parallel, to leveraging the burden of ensemble climate simulations but also offering alternative solutions in case of shutdown (expected or not) of any of the environments. However, for obvious scientific reasons, it is critical to ensure that ESMs provide reproducible results under changes in computing environment. While reproducibility in a strong sense (bit-for-bit) is in general unfeasible, it can be hoped that results obtained under one computing environment are statistically indistinguishable from those obtained under another environment (reproducibility in a weak sense). Here, we develop a protocol to assess the reproducibility (in a weak sense) of the EC-Earth ESM. Using two versions of that ESM, we present one case of non-reproducibility and one case of reproducibility. The non-reproducible case occurs with the older version of the model and likely finds its origin in the treatment of river runoffs along Antarctic coasts. However, it was not possible to nail the problem further down. By contrast, the more recent version of the model provides reproducible results. The protocol introduced in this note can help users of EC-Earth and, by extensión other ESMs, assessing whether their model can be ported from one HPC environment to another. Our results and experience in this work suggest that the default assumption should be that an ESM is not reproducible under changes in the HPC environment, until proven otherwise. Report Antarc* Antarctic DIAL@USL-B (Université Saint-Louis, Bruxelles) Antarctic
institution Open Polar
collection DIAL@USL-B (Université Saint-Louis, Bruxelles)
op_collection_id ftunistlouisbrus
language English
topic High Performance Computing
Climate model
Reproducibility
spellingShingle High Performance Computing
Climate model
Reproducibility
Massonnet, François
Ménégoz, Martin
Acosta, Mario
Yepes-Arbós, Xavier
Exarchout, Eleftheria
Doblas-Reyes, Francisco
Reproducibility of an Earth System Model under a Change in Computing Environment
topic_facet High Performance Computing
Climate model
Reproducibility
description Most Earth System Models (ESMs) are running under different high performance computing (HPC) environments. This has several advantages, from allowing different groups to work with the same tool in parallel, to leveraging the burden of ensemble climate simulations but also offering alternative solutions in case of shutdown (expected or not) of any of the environments. However, for obvious scientific reasons, it is critical to ensure that ESMs provide reproducible results under changes in computing environment. While reproducibility in a strong sense (bit-for-bit) is in general unfeasible, it can be hoped that results obtained under one computing environment are statistically indistinguishable from those obtained under another environment (reproducibility in a weak sense). Here, we develop a protocol to assess the reproducibility (in a weak sense) of the EC-Earth ESM. Using two versions of that ESM, we present one case of non-reproducibility and one case of reproducibility. The non-reproducible case occurs with the older version of the model and likely finds its origin in the treatment of river runoffs along Antarctic coasts. However, it was not possible to nail the problem further down. By contrast, the more recent version of the model provides reproducible results. The protocol introduced in this note can help users of EC-Earth and, by extensión other ESMs, assessing whether their model can be ported from one HPC environment to another. Our results and experience in this work suggest that the default assumption should be that an ESM is not reproducible under changes in the HPC environment, until proven otherwise.
author2 UCL - SST/ELI/ELIC - Earth & Climate
format Report
author Massonnet, François
Ménégoz, Martin
Acosta, Mario
Yepes-Arbós, Xavier
Exarchout, Eleftheria
Doblas-Reyes, Francisco
author_facet Massonnet, François
Ménégoz, Martin
Acosta, Mario
Yepes-Arbós, Xavier
Exarchout, Eleftheria
Doblas-Reyes, Francisco
author_sort Massonnet, François
title Reproducibility of an Earth System Model under a Change in Computing Environment
title_short Reproducibility of an Earth System Model under a Change in Computing Environment
title_full Reproducibility of an Earth System Model under a Change in Computing Environment
title_fullStr Reproducibility of an Earth System Model under a Change in Computing Environment
title_full_unstemmed Reproducibility of an Earth System Model under a Change in Computing Environment
title_sort reproducibility of an earth system model under a change in computing environment
publishDate 2018
url http://hdl.handle.net/2078.1/203820
geographic Antarctic
geographic_facet Antarctic
genre Antarc*
Antarctic
genre_facet Antarc*
Antarctic
op_relation boreal:203820
BSC Technical Memorandum
http://hdl.handle.net/2078.1/203820
op_rights info:eu-repo/semantics/openAccess
_version_ 1798835625015640064