<?xml version="1.0" encoding="utf-8"?>
<export-example>
  <doc>
    <id>62395</id>
    <completedYear/>
    <publishedYear>2025</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>1</pageFirst>
    <pageLast>19</pageLast>
    <pageNumber/>
    <edition/>
    <issue/>
    <volume/>
    <type>article</type>
    <publisherName>SAGE Publications</publisherName>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>1</belongsToBibliography>
    <completedDate>--</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Efficiency and scalability of fully-resolved fluid-particle simulations on heterogeneous CPU-GPU architectures</title>
    <abstract language="eng">Current supercomputers often have a heterogeneous architecture using both conventional Central Processing Units (CPUs) and Graphics Processing Units (GPUs). At the same time, numerical simulation tasks frequently involve multiphysics scenarios whose components run on different hardware due to multiple reasons, e.g., architectural requirements, pragmatism, etc. This leads naturally to a software design where different simulation modules are mapped to different subsystems of the heterogeneous architecture. We present a detailed performance analysis for such a hybrid four-way coupled simulation of a fully resolved particle-laden flow. The Eulerian representation of the flow utilizes GPUs, while the Lagrangian model for the particles runs on conventional CPUs. Two characteristic model situations involving dense and dilute particle systems are used as benchmark scenarios. First, a roofline model is employed to predict the node level performance and to show that the lattice-Boltzmann-based Eulerian fluid simulation reaches very good performance on a single GPU. Furthermore, the GPU-GPU communication for a large-scale Eulerian flow simulation results in only moderate slowdowns. This is due to the efficiency of the CUDA-aware MPI communication, combined with the use of communication hiding techniques. On 1024 A100 GPUs, an overall parallel efficiency of up to 71% is achieved. While the flow simulation has good performance characteristics, the integration of the stiff Lagrangian particle system requires frequent CPU-CPU communications that can become a bottleneck, especially when simulating the dense particle system. Additionally, special attention is paid to the CPU-GPU communication overhead since this is essential for coupling the particles to the flow simulation. However, thanks to our problem-aware co-partitioning, the CPU-GPU communication overhead is found to be negligible. As a lesson learned from this development, four criteria are postulated that a hybrid implementation must meet for the efficient use of heterogeneous supercomputers.</abstract>
    <parentTitle language="eng">The International Journal of High Performance Computing Applications</parentTitle>
    <identifier type="doi">10.1177/10943420241313385</identifier>
    <identifier type="urn">urn:nbn:de:kobv:b43-623958</identifier>
    <identifier type="issn">1741-2846</identifier>
    <enrichment key="opus_doi_flag">true</enrichment>
    <enrichment key="opus_import_data">{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,1,10]],"date-time":"2025-01-10T15:10:23Z","timestamp":1736521823339,"version":"3.32.0"},"reference-count":49,"publisher":"SAGE Publications","license":[{"start":{"date-parts":[[2025,1,10]],"date-time":"2025-01-10T00:00:00Z","timestamp":1736467200000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100031060","name":"European High Performance Computing Joint Undertaking","doi-asserted-by":"crossref","award":["101093393"],"id":[{"id":"10.13039\/100031060","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100001659","name":"Deutsche Forschungsgemeinschaft","doi-asserted-by":"publisher","award":["433735254"],"id":[{"id":"10.13039\/501100001659","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["journals.sagepub.com"],"crossmark-restriction":true},"short-container-title":["The International Journal of High Performance Computing Applications"],"abstract":"&lt;jats:p&gt; Current supercomputers often have a heterogeneous architecture using both conventional Central Processing Units (CPUs) and Graphics Processing Units (GPUs). At the same time, numerical simulation tasks frequently involve multiphysics scenarios whose components run on different hardware due to multiple reasons, e.g., architectural requirements, pragmatism, etc. This leads naturally to a software design where different simulation modules are mapped to different subsystems of the heterogeneous architecture. We present a detailed performance analysis for such a hybrid four-way coupled simulation of a fully resolved particle-laden flow. The Eulerian representation of the flow utilizes GPUs, while the Lagrangian model for the particles runs on conventional CPUs. Two characteristic model situations involving dense and dilute particle systems are used as benchmark scenarios. First, a roofline model is employed to predict the node level performance and to show that the lattice-Boltzmann-based Eulerian fluid simulation reaches very good performance on a single GPU. Furthermore, the GPU-GPU communication for a large-scale Eulerian flow simulation results in only moderate slowdowns. This is due to the efficiency of the CUDA-aware MPI communication, combined with the use of communication hiding techniques. On 1024 A100 GPUs, an overall parallel efficiency of up to 71% is achieved. While the flow simulation has good performance characteristics, the integration of the stiff Lagrangian particle system requires frequent CPU-CPU communications that can become a bottleneck, especially when simulating the dense particle system. Additionally, special attention is paid to the CPU-GPU communication overhead since this is essential for coupling the particles to the flow simulation. However, thanks to our problem-aware co-partitioning, the CPU-GPU communication overhead is found to be negligible. As a lesson learned from this development, four criteria are postulated that a hybrid implementation must meet for the efficient use of heterogeneous supercomputers. &lt;\/jats:p&gt;","DOI":"10.1177\/10943420241313385","type":"journal-article","created":{"date-parts":[[2025,1,10]],"date-time":"2025-01-10T14:34:01Z","timestamp":1736519641000},"update-policy":"https:\/\/doi.org\/10.1177\/sage-journals-update-policy","source":"Crossref","is-referenced-by-count":0,"title":["Efficiency and scalability of fully-resolved fluid-particle simulations on heterogeneous CPU-GPU architectures"],"prefix":"10.1177","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9631-7349","authenticated-orcid":false,"given":"Samuel","family":"Kemmler","sequence":"first","affiliation":[{"name":"Chair for System Simulation, Friedrich-Alexander-Universit\u00e4t Erlangen-N\u00fcrnberg, Erlangen, Germany"},{"name":"Division 7.2 for Buildings and Structures, Bundesanstalt f\u00fcr Materialforschung und -pr\u00fcfung (BAM), Berlin, Germany"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0605-3731","authenticated-orcid":false,"given":"Christoph","family":"Rettinger","sequence":"additional","affiliation":[{"name":"Chair for System Simulation, Friedrich-Alexander-Universit\u00e4t Erlangen-N\u00fcrnberg, Erlangen, Germany"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8796-8599","authenticated-orcid":false,"given":"Ulrich","family":"R\u00fcde","sequence":"additional","affiliation":[{"name":"Chair for System Simulation, Friedrich-Alexander-Universit\u00e4t Erlangen-N\u00fcrnberg, Erlangen, Germany"},{"name":"CERFACS, Toulouse, France"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2446-8065","authenticated-orcid":false,"given":"Pablo","family":"Cu\u00e9llar","sequence":"additional","affiliation":[{"name":"Division 7.2 for Buildings and Structures, Bundesanstalt f\u00fcr Materialforschung und -pr\u00fcfung (BAM), Berlin, Germany"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6992-2690","authenticated-orcid":false,"given":"Harald","family":"K\u00f6stler","sequence":"additional","affiliation":[{"name":"Chair for System Simulation, Friedrich-Alexander-Universit\u00e4t Erlangen-N\u00fcrnberg, Erlangen, Germany"}]}],"member":"179","published-online":{"date-parts":[[2025,1,10]]},"reference":[{"key":"bibr1-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1016\/j.camwa.2020.01.007"},{"key":"bibr2-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1016\/j.jocs.2020.101269"},{"key":"bibr3-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1016\/j.compgeo.2019.103404"},{"key":"bibr4-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcp.2017.03.035"},{"key":"bibr5-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevE.92.053012"},{"key":"bibr6-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1680\/geot.1979.29.1.47"},{"key":"bibr7-10943420241313385","unstructured":"Eibl S, R\u00fcde U (2019) A modular and extensible software architecture for particle dynamics."},{"key":"bibr8-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1016\/j.jpdc.2022.11.003"},{"key":"bibr9-10943420241313385","volume-title":"Design and Performance Evaluation of a Software Framework for Multi-Physics Simulations on Heterogeneous Supercomputers","author":"Feichtinger C","year":"2012"},{"key":"bibr10-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1016\/j.parco.2014.12.003"},{"key":"bibr11-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1016\/j.compgeo.2021.104234"},{"key":"bibr12-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1201\/EBK1439811924"},{"key":"bibr13-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1023\/B:JOSS.0000015179.12689.e4"},{"key":"bibr14-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1016\/j.ces.2020.115712"},{"key":"bibr15-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1137\/22M1531348"},{"key":"bibr16-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1177\/10943420211016525"},{"key":"bibr17-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcp.2024.113337"},{"key":"bibr18-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1007\/s00450-010-0114-4"},{"key":"bibr19-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1016\/j.ces.2013.05.014"},{"key":"bibr20-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1108\/EC-02-2016-0052"},{"key":"bibr21-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1109\/SBGAMES.2010.25"},{"key":"bibr22-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1177\/10943420231158616"},{"key":"bibr23-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1017\/jfm.2014.284"},{"key":"bibr24-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1145\/2304576.2304623"},{"key":"bibr25-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1098\/rsfs.2019.0116"},{"key":"bibr26-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-44649-3"},{"key":"bibr27-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1016\/j.camwa.2009.08.052"},{"key":"bibr28-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1023\/A:1010414013942"},{"key":"bibr29-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevE.106.015308"},{"key":"bibr30-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1016\/j.jocs.2021.101425"},{"key":"bibr31-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1177\/10943420241252511"},{"key":"bibr32-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevE.97.052902"},{"key":"bibr33-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1142\/S0129183198001084"},{"key":"bibr34-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1016\/j.powtec.2016.11.061"},{"key":"bibr35-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1080\/10618562.2017.1390084"},{"key":"bibr36-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1209\/0295-5075\/17\/6\/001"},{"key":"bibr37-10943420241313385","volume-title":"Fully Resolved Simulation of Particulate Flows with a Parallel Coupled Lattice Boltzmann and Discrete Element Method","author":"Rettinger C","year":"2023"},{"key":"bibr38-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1016\/j.compfluid.2017.05.033"},{"key":"bibr39-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcp.2022.110942"},{"key":"bibr40-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-58667-0_1"},{"key":"bibr41-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1017\/jfm.2021.870"},{"key":"bibr42-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1109\/HPCC.2014.14"},{"key":"bibr43-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1017\/jfm.2023.262"},{"key":"bibr44-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevE.91.053301"},{"key":"bibr45-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1109\/CLUSTER.2017.97"},{"key":"bibr46-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1016\/j.powtec.2019.08.095"},{"key":"bibr47-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1016\/j.advwatres.2014.04.019"},{"key":"bibr48-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1016\/j.camwa.2018.01.002"},{"key":"bibr49-10943420241313385","doi-asserted-by":"publisher","DOI":"10.1016\/j.cej.2012.07.049"}],"container-title":["The International Journal of High Performance Computing Applications"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/10943420241313385","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/full-xml\/10.1177\/10943420241313385","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/10943420241313385","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,1,10]],"date-time":"2025-01-10T14:34:07Z","timestamp":1736519647000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/10.1177\/10943420241313385"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,1,10]]},"references-count":49,"alternative-id":["10.1177\/10943420241313385"],"URL":"https:\/\/doi.org\/10.1177\/10943420241313385","relation":{},"ISSN":["1094-3420","1741-2846"],"issn-type":[{"value":"1094-3420","type":"print"},{"value":"1741-2846","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,1,10]]}}}</enrichment>
    <enrichment key="local_crossrefDocumentType">journal-article</enrichment>
    <enrichment key="opus.source">publish</enrichment>
    <enrichment key="date_peer_review">27.01.2025</enrichment>
    <licence>Creative Commons - CC BY - Namensnennung 4.0 International</licence>
    <author>Samuel Kemmler</author>
    <author>Christoph Rettinger</author>
    <author>Ulrich Rüde</author>
    <author>Pablo Cuéllar</author>
    <author>Harald Köstler</author>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Discrete element method</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Hybrid implementation</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>High-performance computing</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Particulate flow</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Lattice Boltzmann method</value>
    </subject>
    <collection role="ddc" number="621">Angewandte Physik</collection>
    <collection role="institutes" number="">7 Bauwerkssicherheit</collection>
    <collection role="institutes" number="">7.2 Ingenieurbau</collection>
    <collection role="themenfelder" number="">Energie</collection>
    <collection role="literaturgattung" number="">Verlagsliteratur</collection>
    <collection role="fulltextaccess" number="">Datei für die Öffentlichkeit verfügbar ("Open Access")</collection>
    <collection role="unnumberedseries" number="">Wissenschaftliche Artikel der BAM</collection>
    <collection role="themenfelder" number="">Windenergie</collection>
    <thesisPublisher>Bundesanstalt für Materialforschung und -prüfung (BAM)</thesisPublisher>
    <file>https://opus4.kobv.de/opus4-bam/files/62395/kemmler-et-al-2025.pdf</file>
  </doc>
  <doc>
    <id>63379</id>
    <completedYear/>
    <publishedYear>2025</publishedYear>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst>1</pageFirst>
    <pageLast>17</pageLast>
    <pageNumber/>
    <edition/>
    <issue/>
    <volume>186</volume>
    <type>article</type>
    <publisherName>Elsevier B.V.</publisherName>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>1</belongsToBibliography>
    <completedDate>--</completedDate>
    <publishedDate>--</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">A fully-resolved micromechanical simulation of piping erosion during a suction bucket installation</title>
    <abstract language="eng">Granular fluidization phenomena such as piping erosion represent a challenge to the delicate installation process of offshore suction bucket foundations. A detailed analysis of the complex conditions in terms of soil composition, soil state, and foundation installation parameters that may lead to piping can be very demanding, if at all possible, solely by experimental means or using macroscopic continuum-based seabed models. The present paper presents a fully-resolved fluid-coupled micromechanical approach for a three-dimensional numerical simulation of the installation process of a suction bucket using the lattice Boltzmann method and discrete element method. The developed model is validated using well-established benchmarks and calibrated by means of experimental data from physical model tests on relevant scenarios focusing on the local fluidization of fixed embedded suction buckets as well as on the suction-driven installation of unrestrained buckets. The qualitative and quantitative agreement with the experimental data both endorse the proposed methodology and highlight the physical soundness of the obtained results. Thereby, the paper shows that three-dimensional analyses of relevant local scenarios at a real scale with little macromechanical model assumptions are feasible.</abstract>
    <parentTitle language="eng">Computers and Geotechnics</parentTitle>
    <identifier type="issn">0266-352X</identifier>
    <identifier type="doi">10.1016/j.compgeo.2025.107375</identifier>
    <identifier type="urn">urn:nbn:de:kobv:b43-633790</identifier>
    <enrichment key="opus_doi_flag">true</enrichment>
    <enrichment key="local_crossrefDocumentType">journal-article</enrichment>
    <enrichment key="local_crossrefLicence">https://www.elsevier.com/tdm/userlicense/1.0/</enrichment>
    <enrichment key="local_import_origin">crossref</enrichment>
    <enrichment key="opus.source">doi-import</enrichment>
    <enrichment key="date_peer_review">30.06.2025</enrichment>
    <licence>Creative Commons - CC BY - Namensnennung 4.0 International</licence>
    <author>Samuel Kemmler</author>
    <author>Pablo Cuéllar</author>
    <author>Antoni Artinov</author>
    <author>Li-Hua Luu</author>
    <author>Abbas Farhat</author>
    <author>Pierre Philippe</author>
    <author>Christoph Rettinger</author>
    <author>Harald Köstler</author>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Micromechanical simulation</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Fluid–solid coupling</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Piping erosion</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Suction bucket foundation</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Offshore wind support structure</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>High-performance computing</value>
    </subject>
    <collection role="ddc" number="621">Angewandte Physik</collection>
    <collection role="institutes" number="">7 Bauwerkssicherheit</collection>
    <collection role="institutes" number="">7.2 Ingenieurbau</collection>
    <collection role="themenfelder" number="">Energie</collection>
    <collection role="literaturgattung" number="">Verlagsliteratur</collection>
    <collection role="fulltextaccess" number="">Datei für die Öffentlichkeit verfügbar ("Open Access")</collection>
    <collection role="unnumberedseries" number="">Wissenschaftliche Artikel der BAM</collection>
    <collection role="themenfelder" number="">Windenergie</collection>
    <thesisPublisher>Bundesanstalt für Materialforschung und -prüfung (BAM)</thesisPublisher>
    <file>https://opus4.kobv.de/opus4-bam/files/63379/1-s2.0-S0266352X25003246-main.pdf</file>
  </doc>
</export-example>
