Last week, members of the European Union CoE RAISE project They gathered at CERN for an ‘all hands’ meeting. This innovative project is developing artificial intelligence (AI) approaches for the next generation of “exascale” supercomputers, for use in both science and industry. Use cases explored through the project include optimizing wind farm layouts, designing efficient aircraft, optimizing acoustic engineering, seismic imaging using remote sensing, and more.
CoE RAISE – European Center of Excellence in Exascale Computing “Research in Artificial Intelligence and Simulation-Based Engineering at Exascale” – is funded under the European Union’s Horizon 2020 Research and Innovation programme. The project was launched in 2021 and will last for three years.
The four-day meeting, which was held in the CERN boardroom, was attended by 54 project members. Participants discussed the progress of their work to develop AI technologies for complex applications in Europe running on future high-performance computing (HPC) systems. Exascale refers to the next generation of high-performance computers that can perform more than 10 machines18 Floating point operations per second (FLOPS). Today only Frontier supercomputer At Oak Ridge National Laboratory in the United States it reached this level. However, with more exascale HPC systems on the horizon, it is important to ensure that AI approaches used in science and industry are ready to fully exploit the huge potential. In June, the European Joint Undertaking for High-Performance Computing (EuroHPC JU) announced that Germany’s Forschungszentrum Jülich GmbH It was chosen to host and operate Europe’s first exascale supercomputerwhich is set to come online next year and will be known as JUPITER (Joint Undertaking Leader for Innovative and Transformational Exascale Research).
CoE RAISE develops innovative approaches to AI on heterogeneous HPC architectures that include multiple processor types. Such architectures can provide higher performance and energy efficiency, but the code must be adapted to efficiently use different types of processors. The artificial intelligence methods being developed focus around it Nine major use cases Well extensible and designed to run on exascale HPC systems.
CoE RAISE supports technology transfer to industry, particularly small and medium-sized businesses, as well as manages education and training initiatives. Furthermore, CoE RAISE also provides consultancy and communicates with other European initiatives to maximize synergies, exploit opportunities for co-design and knowledge sharing. All aspects of the project’s work were discussed over the four days at CERN.
CERN is also a partner and is providing a use case to the project. This work focuses on improving methods for reconstructing particle collision events at the High Luminosity Large Hadron Collider (HL-LHC), which is scheduled to begin operation in 2029. The HL-LHC will see more particle collisions than ever before. Whereas, exabytes of data are generated every year, resulting in unprecedented computing challenges. To reconstruct particle collision events today (with datasets on the order of terabytes or petabytes), hundreds of different algorithms are working concurrently: some are traditional algorithms optimized for specific hardware configurations, while others already incorporate AI-driven methods, such as deep neural networks ( DNNs). Project team members at CERN are working to increase the modularity of the systems and ensure that the code is optimized to take full advantage of heterogeneous architectures, as well as increase the use of machine learning and other AI methods to reconstruct collisions and classify particles.
“Supercomputers are reaching exascale and enabling the delivery of an unprecedented scale of processing resources to HPC and AI workflows,” says Maria Geron, Chief Technology Officer of CERN openlab, who is leading CERN’s contribution to the project. “The research conducted at CoE RAISE will lead to the joint design of HPC computing resources for future AI and HPC applications for both science and industry. This meeting enabled us to exchange ideas, develop them, and introduce new frontiers. It also provided researchers from other fields with unique insight into the environment and challenges facing CERN, which promotes cross-fertilization and understanding.”