Korali: Efficient and scalable software framework for Bayesian uncertainty quantification and stochastic optimization
dc.contributor.author
Martin, Sergio Miguel
dc.contributor.author
Wälchli, Daniel
dc.contributor.author
Arampatzis, Georgios
dc.contributor.author
Economides, Athena E.
dc.contributor.author
Karnakov, Petr
dc.contributor.author
Koumoutsakos, Petros
dc.date.accessioned
2022-12-05T15:03:04Z
dc.date.available
2021-11-29T14:27:14Z
dc.date.available
2021-11-30T10:42:24Z
dc.date.available
2022-01-24T09:50:09Z
dc.date.available
2022-12-05T15:03:04Z
dc.date.issued
2022-02-01
dc.identifier.issn
0045-7825
dc.identifier.issn
1879-2138
dc.identifier.other
10.1016/j.cma.2021.114264
en_US
dc.identifier.uri
http://hdl.handle.net/20.500.11850/517584
dc.identifier.doi
10.3929/ethz-b-000517584
dc.description.abstract
We present Korali, an open-source framework for large-scale Bayesian uncertainty quantification and stochastic optimization. The framework relies on non-intrusive sampling of complex multiphysics models and enables their exploitation for optimization and decision-making. In addition, its distributed sampling engine makes efficient use of massively-parallel architectures while introducing novel fault tolerance and load balancing mechanisms. We demonstrate these features by interfacing Korali with existing high-performance software such as Aphros, LAMMPS (CPU-based), and Mirheo (GPU-based) and show efficient scaling for up to 512 nodes of the CSCS Piz Daint supercomputer. Finally, we present benchmarks demonstrating that Korali outperforms related state-of-the-art software frameworks.
en_US
dc.format
application/pdf
en_US
dc.language.iso
en
en_US
dc.publisher
Elsevier
en_US
dc.rights.uri
http://creativecommons.org/licenses/by/4.0/
dc.subject
High-performance computing
en_US
dc.subject
Bayesian uncertainty quantification
en_US
dc.subject
Optimization
en_US
dc.title
Korali: Efficient and scalable software framework for Bayesian uncertainty quantification and stochastic optimization
en_US
dc.type
Journal Article
dc.rights.license
Creative Commons Attribution 4.0 International
dc.date.published
2021-11-29
ethz.journal.title
Computer Methods in Applied Mechanics and Engineering
ethz.journal.volume
389
en_US
ethz.journal.abbreviated
Comput. Methods Appl. Mech. Eng.
ethz.pages.start
114264
en_US
ethz.size
16 p.
en_US
ethz.version.deposit
publishedVersion
en_US
ethz.grant
Fluid Mechanics in Collective Behaviour: Multiscale Modelling and Applications
en_US
ethz.identifier.wos
ethz.identifier.scopus
ethz.publication.place
Amsterdam
en_US
ethz.publication.status
published
en_US
ethz.leitzahl
ETH Zürich::00002 - ETH Zürich::00012 - Lehre und Forschung::00007 - Departemente::02130 - Dep. Maschinenbau und Verfahrenstechnik / Dep. of Mechanical and Process Eng.::03499 - Koumoutsakos, Petros (ehemalig) / Koumoutsakos, Petros (former)
en_US
ethz.leitzahl.certified
ETH Zürich::00002 - ETH Zürich::00012 - Lehre und Forschung::00007 - Departemente::02130 - Dep. Maschinenbau und Verfahrenstechnik / Dep. of Mechanical and Process Eng.::03499 - Koumoutsakos, Petros (ehemalig) / Koumoutsakos, Petros (former)
en_US
ethz.grant.agreementno
341117
ethz.grant.fundername
EC
ethz.grant.funderDoi
10.13039/501100000780
ethz.grant.program
FP7
ethz.date.deposited
2021-11-29T14:27:35Z
ethz.source
FORM
ethz.eth
yes
en_US
ethz.availability
Open access
en_US
ethz.rosetta.installDate
2022-01-24T09:50:16Z
ethz.rosetta.lastUpdated
2023-02-07T08:28:58Z
ethz.rosetta.versionExported
true
ethz.COinS
ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.atitle=Korali:%20Efficient%20and%20scalable%20software%20framework%20for%20Bayesian%20uncertainty%20quantification%20and%20stochastic%20optimization&rft.jtitle=Computer%20Methods%20in%20Applied%20Mechanics%20and%20Engineering&rft.date=2022-02-01&rft.volume=389&rft.spage=114264&rft.issn=0045-7825&1879-2138&rft.au=Martin,%20Sergio%20Miguel&W%C3%A4lchli,%20Daniel&Arampatzis,%20Georgios&Economides,%20Athena%20E.&Karnakov,%20Petr&rft.genre=article&rft_id=info:doi/10.1016/j.cma.2021.114264&
Files in this item
Publication type
-
Journal Article [130918]