AI and supercomputers join forces to tackle global challenges
There is immense potential in combining AI technology and high-performance computing (HPC) to address the most complex global challenges that are beyond existing capabilities. This is where exascale computing comes in to provide much more speed and power than any of today’s most advanced supercomputers. However, there are several challenges to integrating high computational power and scalable AI solutions in the analysis and processing of big data. One of the ways the EU-funded RAISE project is addressing this issue is to develop novel AI methods on heterogeneous HPC architectures that provide superior performance and energy efficiency. The AI methods are focused on several use cases and intended to scale well for running on exascale HPC systems. These use cases cover a broad range of key academic and industrial applications, such as wind energy. Researchers proposed developing machine learning-trained surrogate models of wind turbines to analyse their impact on each other and their interaction with the wind farm topography. To improve the accuracy of the wind farm simulations and enhance their computational performance, they also proposed and validated a novel meshing approach, where the turbines were modelled with actuator discs. This wind farm simulation framework enables mesh adaptation to the solution and has the capacity to align the disc models with the flow. As a result, simulations of complex wind farm configurations can be accurately carried out. The findings were published in the peer-reviewed journal ‘Energies’.
Supporting tech transfer and knowledge sharing
RAISE, a ‘Centres of Excellence’ (CoE) undertaking, is bringing together AI and HPC experts to ensure knowledge and technology transfer. In January 2023, 54 consortium members gathered at project partner European Organization for Nuclear Research (CERN) to present and discuss the use cases. “The participants in the meeting … initiated more intensified collaborations on emerging topics, such as the integration of quantum computing in AI and HPC workflows,” comments Andreas Lintermann of project coordinator Forschungszentrum Jülich GmbH, Germany. “Supercomputers are reaching the exascale and enabling the delivery of an unprecedented scale of processing resources for HPC and AI workflows,” adds Maria Girone of CERN. “The research performed in CoE RAISE will drive the co-design of HPC computing resources for future AI and HPC applications for both science and industry. This meeting enabled us to exchange and develop ideas and to bring new perspectives.” Ending in December 2023, the RAISE (Research on AI- and Simulation-Based Engineering at Exascale) project is developing AI approaches for exascale supercomputers that will be used in both academia and industry. If you are interested in having your project featured as a ‘Project of the Month’ in an upcoming issue, please send us an email to editorial@cordis.europa.eu and tell us why!
Keywords
RAISE, computer, supercomputer, AI, exascale, high-performance computing, HPC