001049804 001__ 1049804 001049804 005__ 20260108204824.0 001049804 0247_ $$2doi$$a10.22323/1.466.0280 001049804 0247_ $$2datacite_doi$$a10.34734/FZJ-2025-05588 001049804 037__ $$aFZJ-2025-05588 001049804 041__ $$aEnglish 001049804 1001_ $$0P:(DE-HGF)0$$aAltherr, Anian$$b0 001049804 1112_ $$aThe 41st International Symposium on Lattice Field Theory$$cLiverpool$$d2024-07-28 - 2024-08-03$$gLATTICE 2024$$wUK 001049804 245__ $$aO(a)-improved QCD+QED Wilson Dirac operator on GPUs 001049804 260__ $$c2025 001049804 29510 $$aProceedings of The 41st International Symposium on Lattice Field Theory 001049804 300__ $$a28 001049804 3367_ $$2ORCID$$aCONFERENCE_PAPER 001049804 3367_ $$033$$2EndNote$$aConference Paper 001049804 3367_ $$2BibTeX$$aINPROCEEDINGS 001049804 3367_ $$2DRIVER$$aconferenceObject 001049804 3367_ $$2DataCite$$aOutput Types/Conference Paper 001049804 3367_ $$0PUB:(DE-HGF)8$$2PUB:(DE-HGF)$$aContribution to a conference proceedings$$bcontrib$$mcontrib$$s1767859784_29748 001049804 3367_ $$0PUB:(DE-HGF)7$$2PUB:(DE-HGF)$$aContribution to a book$$mcontb 001049804 4900_ $$aProceedings of Science 001049804 520__ $$aMarkov Chain Monte Carlo simulations of lattice Quantum Chromodynamics (QCD) are the onlyknown tool to investigate non-perturbatively the theory of the strong interaction and are requiredto perform precision tests of the Standard Model of Particle Physics. As the Markov Chain isa serial process, the sole option for improving the sampling rate is accelerating each individualupdate step. Heterogeneous clusters of GPU-accelerated nodes offer large total memory bandwidthwhich can be used to speed-up our application, openQxD-1.1, which is dominated by inversionsof the Dirac operator, a large sparse matrix. In this work we investigate offloading the inversionto GPU using the lattice-QCD library QUDA, and our early results demonstrate a significantpotential speed-up in the time-to-solution for state-of-the-art problem sizes. Minimal extensionsto the existing QUDA library are required for our specific physics programme while greatlyenhancing the performance portability of our code and retaining the reliability and robustness ofexisting applications in openQxD-1.1. Our new interface will enable us to utilize pre-exascaleinfrastructure and reduce the systematic uncertainty in our physics predictions by incorporatingthe effects of quantum electromagnetism (QED) in our simulations. 001049804 536__ $$0G:(DE-HGF)POF4-5111$$a5111 - Domain-Specific Simulation & Data Life Cycle Labs (SDLs) and Research Groups (POF4-511)$$cPOF4-511$$fPOF IV$$x0 001049804 588__ $$aDataset connected to CrossRef Conference 001049804 7001_ $$0P:(DE-HGF)0$$aCampos, Isabel$$b1 001049804 7001_ $$0P:(DE-HGF)0$$aColes, Jonathan$$b2 001049804 7001_ $$0P:(DE-Juel1)207074$$aCotellucci, Alessandro$$b3$$ufzj 001049804 7001_ $$0P:(DE-HGF)0$$aFernández De la Garza, Juan Antonio$$b4 001049804 7001_ $$0P:(DE-HGF)0$$aGruber, Roman$$b5$$eCorresponding author 001049804 7001_ $$0P:(DE-HGF)0$$aHarris, Tim$$b6 001049804 7001_ $$0P:(DE-HGF)0$$aKomijani, Javad$$b7 001049804 7001_ $$0P:(DE-HGF)0$$aLücke, Jens$$b8 001049804 7001_ $$0P:(DE-HGF)0$$aMaier, Stephanie$$b9 001049804 7001_ $$0P:(DE-HGF)0$$aMarinkovic, Marina$$b10 001049804 7001_ $$0P:(DE-HGF)0$$aParato, Letizia$$b11 001049804 7001_ $$0P:(DE-HGF)0$$aPatella, Agostino$$b12 001049804 7001_ $$0P:(DE-HGF)0$$aRosso, Sara$$b13 001049804 7001_ $$0P:(DE-HGF)0$$aTavella, Paola$$b14 001049804 7001_ $$0P:(DE-HGF)0$$aVogt, Hannes$$b15 001049804 773__ $$a10.22323/1.466.0280$$y2025 001049804 8564_ $$uhttps://juser.fz-juelich.de/record/1049804/files/LATTICE2024_280.pdf$$yOpenAccess 001049804 909CO $$ooai:juser.fz-juelich.de:1049804$$popenaire$$popen_access$$pVDB$$pdriver$$pdnbdelivery 001049804 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)207074$$aForschungszentrum Jülich$$b3$$kFZJ 001049804 9131_ $$0G:(DE-HGF)POF4-511$$1G:(DE-HGF)POF4-510$$2G:(DE-HGF)POF4-500$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$9G:(DE-HGF)POF4-5111$$aDE-HGF$$bKey Technologies$$lEngineering Digital Futures – Supercomputing, Data Management and Information Security for Knowledge and Action$$vEnabling Computational- & Data-Intensive Science and Engineering$$x0 001049804 9141_ $$y2025 001049804 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess 001049804 915__ $$0LIC:(DE-HGF)CCBYNCND4$$2HGFVOC$$aCreative Commons Attribution-NonCommercial-NoDerivs CC BY-NC-ND 4.0 001049804 9201_ $$0I:(DE-Juel1)JSC-20090406$$kJSC$$lJülich Supercomputing Center$$x0 001049804 980__ $$acontrib 001049804 980__ $$aVDB 001049804 980__ $$aUNRESTRICTED 001049804 980__ $$acontb 001049804 980__ $$aI:(DE-Juel1)JSC-20090406 001049804 9801_ $$aFullTexts