This article was accepted into the corpus but its outbound wikilinks were never NER-processed — typical at the deepest BFS hop or when the run's entity cap was reached. No expansion funnel to show.
| PASC (Platform for Advanced Scientific Computing) | |
|---|---|
| Name | PASC (Platform for Advanced Scientific Computing) |
| Formation | 2007 |
| Headquarters | Zurich, Switzerland |
| Region served | Europe |
| Leader title | Director |
| Parent organization | Swiss National Supercomputing Centre |
PASC (Platform for Advanced Scientific Computing) is a European research platform that fosters high-performance computing collaboration among universities, research centers, and industry. It supports computational science, numerical analysis, and simulation projects by coordinating resources, funding, and training across institutions. PASC acts as a nexus linking national laboratories, supercomputing centers, and academic departments to accelerate computational methods for grand-challenge problems.
PASC connects organizations such as Swiss National Supercomputing Centre, ETH Zurich, EPFL, Max Planck Society, and CERN with partners like IBM, NVIDIA, Intel Corporation, and Microsoft Research to advance simulations in physics, chemistry, and engineering. The platform promotes collaborations with institutions including University of Cambridge, ETH Zurich Department of Mathematics, University of Oxford, Technical University of Munich, and École Polytechnique Fédérale de Lausanne to develop scalable algorithms, software stacks, and middleware. PASC emphasizes interoperability with facilities like PRACE, EuroHPC, National Center for Supercomputing Applications, and Lawrence Berkeley National Laboratory while engaging consortia such as HPC Wales, Barcelona Supercomputing Center, and Leibniz Supercomputing Centre.
PASC emerged from initiatives led by entities like Swiss National Science Foundation, European Research Council, ETH Board, and academic groups at ETH Zurich and EPFL to coordinate Swiss and European high-performance computing efforts. Early milestones involved pilots with IBM Blue Gene, collaborations with Cray Inc., benchmarking against systems at Oak Ridge National Laboratory and Argonne National Laboratory, and project funding from Horizon 2020 and bilateral grants with German Research Foundation. Key advisors and contributors included researchers affiliated with University of Zurich, University of Basel, University of Geneva, and visiting scholars from Princeton University and Massachusetts Institute of Technology.
PASC promotes software ecosystems integrating libraries and frameworks such as MPI, OpenMP, CUDA, Kokkos, PETSc, Trilinos, HDF5, and Zoltan deployed on hardware from NVIDIA, AMD, Intel Corporation, Arm Holdings, and systems vendors like Hewlett Packard Enterprise and Lenovo. The platform's stack interoperates with job schedulers and resource managers like Slurm Workload Manager, PBS Professional, and IBM Spectrum LSF and leverages filesystems such as Lustre and GPFS. PASC-supported environments include container and workflow tools from Docker, Singularity (software), Kubernetes, and workflow engines used at Swiss National Supercomputing Centre and Barcelona Supercomputing Center. Compiler toolchains from GNU Compiler Collection, LLVM, and proprietary compilers from Intel Corporation are integrated with performance tools like VTune, TAU (software), Score-P, and profiling suites from NVIDIA.
Research under PASC spans computational fluid dynamics applied by groups at Imperial College London and Delft University of Technology, climate modeling used by teams at Met Office and ETH Zurich, computational chemistry advanced by researchers at Max Planck Institute for Polymer Research and University of Cambridge, and materials science pursued with partners such as CERN and Paul Scherrer Institute. Application domains include astrophysics collaborations with European Southern Observatory, biomedical simulations with Swiss Federal Institute of Technology in Lausanne, and energy systems modeling involving École Polytechnique and TU Delft. Projects have produced scalable solvers, reduced-order models, and machine-learning-augmented simulations integrating methods from groups at University of California, Berkeley, Stanford University, and Carnegie Mellon University.
PASC is funded through mechanisms involving Swiss National Science Foundation, Horizon 2020, European Commission, industry partnerships with Siemens, ABB, Shell, and collaborative grants with national agencies such as SNSF and Swedish Research Council. Strategic partnerships include joint activities with PRACE, EuroHPC, European Molecular Biology Laboratory, and bilateral research programs with German Research Foundation and French National Centre for Scientific Research. Funding programs and calls often reference frameworks developed in concert with ETH Board and regional innovation agencies.
Governance of the platform involves advisory boards and steering committees populated by representatives from Swiss National Supercomputing Centre, ETH Zurich, EPFL, University of Zurich, and partner institutions like Max Planck Society and CERN. Community-building efforts include workshops co-located with conferences such as Supercomputing Conference, ISC High Performance, SIAM Conference on Computational Science and Engineering, and summer schools modeled after programs at Los Alamos National Laboratory and Lawrence Livermore National Laboratory. PASC maintains working groups focused on software sustainability, reproducible research, and workforce development linked to doctoral programs at ETH Zurich and postdoctoral networks supported by Marie Skłodowska-Curie Actions.
PASC evaluates software and hardware using benchmarks derived from real-world applications and standardized suites used by TOP500, SPEC, and community benchmarks from PRACE and EuroHPC. Performance studies compare implementations on testbeds provided by Swiss National Supercomputing Centre, CSCS, Leibniz Supercomputing Centre, and partner centers such as Jülich Research Centre. Metrics include scaling efficiencies reported at venues like Supercomputing Conference and validated through collaborations with research teams at Argonne National Laboratory, Oak Ridge National Laboratory, and industrial partners including NVIDIA and Intel Corporation.
Category:Scientific organizations Category:High-performance computing