001026685 001__ 1026685 001026685 005__ 20250822121412.0 001026685 0247_ $$2doi$$a10.5281/ZENODO.10214076 001026685 037__ $$aFZJ-2024-03500 001026685 041__ $$aEnglish 001026685 1001_ $$00000-0002-5699-1793$$aGarcia de Gonzalo, Simon$$b0 001026685 1112_ $$aThe International Conference for High Performance Computing, Networking, Storage, and Analysis 2023$$cDenver, CO$$d2023-11-12 - 2023-11-17$$gSC23$$wUSA 001026685 245__ $$aEfficient Distributed GPU Programming for Exascale 001026685 260__ $$c2023 001026685 3367_ $$033$$2EndNote$$aConference Paper 001026685 3367_ $$2DataCite$$aOther 001026685 3367_ $$2BibTeX$$aINPROCEEDINGS 001026685 3367_ $$2DRIVER$$aconferenceObject 001026685 3367_ $$2ORCID$$aLECTURE_SPEECH 001026685 3367_ $$0PUB:(DE-HGF)6$$2PUB:(DE-HGF)$$aConference Presentation$$bconf$$mconf$$s1717751105_3800$$xAfter Call 001026685 520__ $$aOver the past years, GPUs became ubiquitous in HPC installations around the world, delivering the majority of performance of some of the largest supercomputers (e.g. Summit, Sierra, JUWELS Booster). This trend continues in the recently deployed and upcoming Pre-Exascale and Exascale systems (LUMI, Leonardo; Frontier, Perlmutter): GPUs are chosen as the core computing devices to enter this next era of HPC. To take advantage of future GPU-accelerated systems with tens of thousands of devices, application developers need to have the propers skills and tools to understand, manage, and optimize distributed GPU applications. In this tutorial, participants will learn techniques to efficiently program large-scale multi-GPU systems. While programming multiple GPUs with MPI is explained in detail, also advanced tuning techniques and complementing programming models like NCCL and NVSHMEM are presented. Tools for analysis are shown and used to motivate and implement performance optimizations. The tutorial teaches fundamental concepts that apply to GPU-accelerated systems in general, taking the NVIDIA platform as an example. It is a combination of lectures and hands-on exercises, using one of Europe's fastest supercomputers, JUWELS Booster, for interactive learning and discovery. 001026685 536__ $$0G:(DE-HGF)POF4-5121$$a5121 - Supercomputing & Big Data Facilities (POF4-512)$$cPOF4-512$$fPOF IV$$x0 001026685 536__ $$0G:(DE-HGF)POF4-5122$$a5122 - Future Computing & Big Data Systems (POF4-512)$$cPOF4-512$$fPOF IV$$x1 001026685 536__ $$0G:(DE-HGF)POF4-5112$$a5112 - Cross-Domain Algorithms, Tools, Methods Labs (ATMLs) and Research Groups (POF4-511)$$cPOF4-511$$fPOF IV$$x2 001026685 536__ $$0G:(DE-Juel-1)ATML-X-DEV$$aATML-X-DEV - ATML Accelerating Devices (ATML-X-DEV)$$cATML-X-DEV$$x3 001026685 588__ $$aDataset connected to DataCite 001026685 7001_ $$0P:(DE-Juel1)145478$$aHerten, Andreas$$b1$$eCorresponding author 001026685 7001_ $$0P:(DE-HGF)0$$aHrywniak, Markus$$b2 001026685 7001_ $$0P:(DE-HGF)0$$aKraus, Jiri$$b3 001026685 7001_ $$0P:(DE-Juel1)188270$$aOden, Lena$$b4 001026685 773__ $$a10.5281/ZENODO.10214076 001026685 8564_ $$uhttps://github.com/FZJ-JSC/tutorial-multi-gpu/tree/v5.0-sc23 001026685 909CO $$ooai:juser.fz-juelich.de:1026685$$pVDB 001026685 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)145478$$aForschungszentrum Jülich$$b1$$kFZJ 001026685 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)188270$$aForschungszentrum Jülich$$b4$$kFZJ 001026685 9131_ $$0G:(DE-HGF)POF4-512$$1G:(DE-HGF)POF4-510$$2G:(DE-HGF)POF4-500$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$9G:(DE-HGF)POF4-5121$$aDE-HGF$$bKey Technologies$$lEngineering Digital Futures – Supercomputing, Data Management and Information Security for Knowledge and Action$$vSupercomputing & Big Data Infrastructures$$x0 001026685 9131_ $$0G:(DE-HGF)POF4-512$$1G:(DE-HGF)POF4-510$$2G:(DE-HGF)POF4-500$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$9G:(DE-HGF)POF4-5122$$aDE-HGF$$bKey Technologies$$lEngineering Digital Futures – Supercomputing, Data Management and Information Security for Knowledge and Action$$vSupercomputing & Big Data Infrastructures$$x1 001026685 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)POF4-5112$$aDE-HGF$$bKey Technologies$$lEngineering Digital Futures – Supercomputing, Data Management and Information Security for Knowledge and Action$$vEnabling Computational- & Data-Intensive Science and Engineering$$x2 001026685 9141_ $$y2024 001026685 920__ $$lyes 001026685 9201_ $$0I:(DE-Juel1)JSC-20090406$$kJSC$$lJülich Supercomputing Center$$x0 001026685 980__ $$aconf 001026685 980__ $$aVDB 001026685 980__ $$aI:(DE-Juel1)JSC-20090406 001026685 980__ $$aUNRESTRICTED