Conference Agenda
Overview and details of the sessions of this conference. Please select a date or location to show only sessions at that day or location. Please select a single session for detailed view (with abstracts and downloads if available).
Please note that all times are shown in the time zone of the conference. The current conference time is: 2nd Sept 2026, 01:53:20am CEST
|
Daily Overview |
| 7:30am - 7:45am |
Transfer from Aachen 07:30 departure: Hotel Motel One Aachen, Kapuzinergraben 6-10, 52062 Aachen 07:45 departure: Novotel Aachen City, Peterstraße 66, 52062 Aachen 07:45 departure: Hotel Ibis Marschiertor, Friedlandstraße 6-8, 52064 Aachen |
||
| 8:40am - 9:00am |
Registration and Orientation Location: Central Library (Foyer) |
||
| 9:00am - 9:15am |
Welcome Location: Central Library (Lecture hall) Support: Leon Thelen |
||
| 9:15am - 10:00am |
Keynote Lecture Satoshi Matsuoka: The Game-changing aspects of AI for Science onto HPC Location: Central Library (Lecture hall) Chair: Kentaro Sano Support: Leon Thelen |
||
| 10:00am - 10:30am |
Walk to JSC and Coffee Break Location: Foyer, JSC |
||
| 10:30am - 12:00pm |
Short Talks Programming Languages Location: Rotunda, JSC Chair: David Álvarez Robert Support: Jannik Finck Integration of scheduling of communications and scheduling of tasks in a distributed task-based runtime system 1: Centre Inria de l'Université de Bordeaux, France; 2: University of Bordeaux, France; 3: Bordeaux-INP, France Agent coordination in the Compute Continuum Barcelona Supercomputing Center, Spain Extending Languages for Productive and Performant Scientific Parallel Programming Argonne National Laboratory, United States of America A common framework for dynamic load balancing based on overdecomposition University of Illinois / NCSA, United States of America Programming Models for Leveraging Programmable SmartNICs for HPC and AI Barcelona Supercomputing Center, Spain |
Short Talks Numerical Methods Location: SR 1, JSC Chair: Jakob Fritz Support: Frederick Löffel diff-LAPACK: A library of differentiated BLAS/LAPACK routines 1: Argonne National Lab, USA; 2: INRIA, France Counterexamples to Payne's nodal line conjecture using HPC Forschungszentrum Jülich, Germany ExaNLA: An Application-Driven Initiative for Exascale Dense Numerical Linear Algebra on Accelerated Architectures Jülich Supercomputing Centre, Forschungszentrum Jülich, Germany Personalized medicine through multi-physics flow simulations Forschungszentrum Juelich, Germany Formal Verification of Scientific Software Argonne National Laboratory, United States of America Two uses of complex analysis for scientific computing applications Argonne National Lab, United States of America |
Break-out Session AI Location: Lecture Hall, JSC Chair: Franck Cappello Support: Oskar Druska Program:
Full version here! HPC Challenges in AI Factories BSC CNS, Spain |
| 12:00pm - 1:30pm |
Lunch Break Location: Central Library (Foyer) |
JSC Tours 1+2 Location: Central Library (Foyer) Support: Oskar Druska (Tour 1), Leon Thelen (Tour 2) |
|
| 1:30pm - 3:00pm |
Short Talks Performance Tools Location: Rotunda, JSC Chair: Ritvik Rao Support: Leon Thelen Pallas: Purrformance analysis for large-scale computing 1: Télécom SudParis, France; 2: ENSIIE, France BRAVAS: Towards a Portable GPU Profiler Barcelona Supercomputing Center, Spain Progress in Continuous Benchmarking at JSC Forschungszentrum Jülich, Germany NRM: An Infrastructure for Dynamic Reconfiguration of Systems Argonne National Laboratory, United States of America Centralised Dashboard for Continuous Benchmarking: From HPC Clusters to Quantum Processors Forschungszentrum Juelich GmbH, Germany Opportunistic LLM Inference on Idle Resources in Extreme-Scale HPC Systems Argonne National Laboratory, United States of America |
Short Talks Advanced Architectures Location: SR 1, JSC Chair: Tomoya Yuki Support: Carina Himmels Plant your virtual trees! A distributed GPU-Native Octree for the FMM Forschungszentrum Jülich, Germany FMM-azing Adventures Beyond PME in Molecular Dynamics Simulations Forschungszentrum Jülich, Germany Computational Storage for High-Performance Computing Inria, France AI-Assisted Hardware Design with Multi-Platform RTL Verification and Physical Feasibility Analysise Argonne National Laboratory, United States of America AI-Assisted Chip Development: Status and Future Directions Argonne National Laboratory, United States of America Dealing with aging HPC Inria, France |
Break-out Session AI Location: Lecture Hall, JSC Chair: Kengo Nakajima Support: Oskar Druska Program:
Full version here! Opportunities to Collaborate on Integrating LLMs with Simulation Software RIKEN R-CCS, Japan |
| 3:00pm - 3:30pm |
Coffee Break Location: Foyer, JSC |
||
| 3:30pm - 5:00pm |
Short Talks I/O Location: Rotunda, JSC Chair: Théo Jolivel Support: Leon Thelen Recent Advances in Error-Bounded Lossy Compression with SZ for Scientific and AI Workloads Argonne National Laboratory, United States of America When Control Theory Meets HPC Workflows: A Data-Driven Approach to Dynamic Optimization Argonne National Laboratory, United States of America Thread concurrent parallel I/O in SIONlib Forschungszentrum Juelich, Germany Scalable Persistence of Artifacts and Reasoning Chains in Agentic Workflows Argonne National Laboratory, United States of America Scalable Data Management Techniques for AI workloads Argonne National Laboratory, United States of America I/O Traces Analysis: What’s Next? Inria, France |
Individual discussions Location: SR 1, JSC Use this slot to get together and discuss new ideas, new problems, collaborations, opportunities, and the likes. |
Break-out Session AI Location: Lecture Hall, JSC Chair: Jordi Atserias Support: Oskar Druska Program:
Full program here! Innovative Computational Science by Integration of Simulation/Data/Learning on Heterogeneous Environments towards FugakuNEXT 1: RIKEN, Japan; 2: The University of Tokyo, Japan; 3: Juelich Supercomputing Center, Germany Evaluating the Relevance of Task-based Runtime Systems to Implement Large Language Model Applications 1: Bordeaux INP; 2: Inria |
| 5:00pm | Walk to Central Library |
||
| 5:15pm - 6:00pm |
Poster Blitz Location: Central Library (Lecture hall) Chair: Robert Speck Support: Leon Thelen AI-Assisted Hardware Design with Multi-Platform RTL Verification and Physical Feasibility Analysise Argonne National Laboratory, United States of America Next Steps in Large-Scale Density Functional Theory Jülich Supercomputing Centre, Forschungszentrum Jülich, Jülich, Germany The Kronos Project: Hybrid Discrete Event Simulations of Computing Infrastructure Argonne National Laboratory (ANL), United States of America Design of Identity and Access Control for the Quantum–HPC Hybrid Platform 1: RIKEN R-CCS; 2: University of Tsukuba; 3: Juntendo University AI-based Sub-Grid Scale Closure for Large Eddy Simulations in the Human Larynx Forschungszentrum Jülich Gmbh, Germany RIKEN-braket: massively parallel simulatior of quantum computers 1: RIKEN Center for Computational Science; 2: Lawrence Berkeley National Laboratory Following aging in HPC systems Univ Rennes, Inria, CNRS, IRISA, France Adaptive-precision interatomic potentials (APIP) 1: Jülich Supercomputing Centre (JSC), Forschungszentrum Jülich, Germany; 2: Interdisciplinary Centre for Advanced Materials Simulations (ICAMS), Ruhr Universität Bochum, German Enhancing Scientific Image Prediction and Compression through AI Model Fine-Tuning 1: Kanazawa University, Japan; 2: RIKEN Center for Computational Science, Japan JUNIQ Benchmark Suite: Tracking Progress in Quantum Technology Readiness Jülich Supercomputing Centre (JSC), Germany ScaFaCoS 2.0: A Performance-Portable Coulomb Solver Library for Exascale Simulations Forschungszentrum Jülich, Germany Composability from numerical algorithms to programming model Inria, France Modeling and reuse of data partitioning code in HPC Inria, France Development of a state vector quantum computer simulator for GPUs Riken Center for Computational Science, Japan Improving cardinality constraints for the annealer using bineary search and adders Forschungszentrum Jülich, Germany Lossy and Lossless Data Compression of Meteorological Data for Lagrangian Transport Modeling Jülich Supercomputing Centre, Forschungszentrum Jülich, Germany An ANARI-Based Pipeline from ParaView to Unreal Engine, Enabled by JuSync for In-Situ HPC Visualization 1: Forschungszentrum Jülich GmbH, Germany; 2: Leibniz-Rechenzentrum, Germany; 3: Argonne National Laboratory, USA Centralised Dashboard for Continuous Benchmarking: From HPC Clusters to Quantum Processors Forschungszentrum Juelich GmbH, Germany Performance Characterization of TSMP2 on Heterogeneous HPC Architectures 1: Forschungszentrum Jülich GmbH, Germany; 2: Technical University of Dresden, Germany JUBE: An Environment for systematic benchmarking and scientific workflows Forschungszentrum Juelich GmbH, Germany Building multi-scale modeling tool for optimizing brain stimulation in Alzheimer's disease 1: Forschungszentrum Jülich, Germany; 2: University Cantabria; 3: Technical University Darmstadt; 4: CY Cergy Paris University A Case Study on Hybrid Quantum-Classical Workflow Modeling 1: Inria Rennes, France; 2: Barcelona Supercomputing Center, Spain; 3: Inria Saclay, France Automated code generation of advanced plasticity rules for the SpiNNaker neuromorphic platform using NESTML 1: Simulation and Data Laboratory Neuroscience, Jülich Supercomputer Centre, Institute for Advanced Simulation, Jülich-Aachen Research Alliance, Forschungszentrum Jülich GmbH; 2: Institute for Advanced Simulation IAS-6, Computational and Systems Neuroscience, Forschungszentrum Jülich GmbH; 3: Software Engineering, Software Engineering, RWTH Aachen University, Germany Modeling and simulating spiking neurons and synaptic plasticity with NESTML on HPC 1: Forschungszentrum Jülich, Germany; 2: Software Engineering, RWTH Aachen University, Germany Plant your virtual trees! A distributed GPU-Native Octree for the FMM Forschungszentrum Jülich, Germany From JUWELS to JUPITER: Scaling ICON Toward Kilometer-Scale Earth System Simulation and Implications for a German Earth System Model Forschungszentrum Jülich GmbH, Germany Controlling the temperature of computing sitesby injecting useful, non-invasive tasks 1: Univ. Grenoble Alpes, Inria, CNRS, Grenoble INP, LIG, Grenoble, France; 2: Univ. Grenoble Alpes, CNRS, Grenoble INP, GIPSA-lab, Grenoble, France; 3: Univ. Lille, Inria, CNRS, Centrale Lille, CRIStAL, Lille, France; 4: Qarnot Computing, Montrouge, France GPUs are Fast, You May be Not Forschungszentrum Jülich, Germany FMM-azing Adventures Beyond PME in Molecular Dynamics Simulations Forschungszentrum Jülich, Germany Performance Evaluation and Optimization of an MPS-Based CFD Solver on GPUs 1: RIKEN Center for Computational Science, Japan; 2: Kobe University, Japan; 3: Hokkaido University, Japan Lessons Learned from Power-Saving Operations on Fugaku: Operational Analysis and Benchmark Validation 1: RIKEN R-CCS, Japan; 2: Fujitsu Ltd. Practical FHE Inference with Parameter Study on GPU Barcelona Super-Computing Center, Spain DPA-CCL: Offloading Collective Communications to Data Path Accelerators Barcelona Supercomputing Center, Spain Performance of Quantum Autoencoders in Time-Series Anomaly Detection Forschungszentrum Jülich, Germany Malleability in hybrid quantum-classical programs: adaptive classical resource allocation across iterative workflows Barcelona Super Computer Center, Spain |
||
| 6:00pm - 7:30pm |
Poster Session and Dinner Location: Central Library (Foyer) |
||
| 7:30pm | Transfer to Aachen |
||