CERFACS participated to the Teratec23 forum on High-Performance computing at the Parc Floral de Paris on May 31st-June 1st, 2023. This forum has welcomed international experts in Super Computing from all sectors of our society, such as Climate and Geosciences. As a leader in the landscape of Digital Simulation, CERFACS has taken part in two workshop sessions.
Gabriel Staffelbach (HPC expert, CERFACS) was involved in “Quantum Computing: Future of HPC?” session. Sophie Ricci (Data assimilation expert, CERFACS/CNRS) and Thanh Huy Nguyen (Remote sensing expert, CERFACS/CNRS) were involved in the “HPC for the Earth Sciences, Climate and Environment” workshop.
Gabriel Staffelbach delivered an insightful presentation on CERFACS’ technology watch initiatives, with a specific focus on harnessing the potential of Quantum computer simulators and Quantum hardware for solving partial differential equations. The presentation highlighted the culmination of collaborative efforts spanning the past three years, involving key partnerships with Eviden (an Atos business) and IBM. In addition to this, CERFACS worked with the LIRMM laboratory to delve into the realm of quantum computing, leveraging their extensive experience in High-Performance Computing (HPC) as a solid foundation. This comprehensive exploration aims to unravel the advantages and drawbacks of quantum computing in the context of solving complex equations.
S. Ricci and T. H. Nguyen gave a joint presentation with Raquel Rodriguez-Suquet (CNES, Lab’OT) on the FloodDAM-DT project.
FloodDAM-DT (Flood Detection Alert and mapping – Digital Twin) is supported by the Space for Climate Observatory (SCO) program as a follow-up to its precursor FloodDAM within an international French-American cooperation. The international cooperation between CNES SCO FloodDAM-DT and NASA IDEAS project (AIST program) on water-cycle applications to devise an Earth System Digital Twin (ESDT) for alert and flood risk mapping systems at small and global scale to enable researchers and decision-makers to visualize, analyze, and evaluate different scenario outcomes for environmental matters where risk assessment is at stake. In this context, the targeted Digital twin architecture for Hydrology aims at dynamically connecting existing hydrologic/hydrodynamic models and continental water data (in-situ, airborne, remote-sensing (RS) data sources e.g. radar and optical) from both agencies (NASA & CNES) to produce large-scale flood predictions. This work entails a multi-agency effort to establish an extensible open-source platform focused on water resources and flood analysis where test case studies serve as demonstrators.
This 18-month project will result in a Proof-of-Concept showcase where the processing blocks are assembled for the Garonne catchment in France and the Ohio catchment in the US, thus providing access to in-situ and satellite/drone data, flood extent mapping issued from data processing, and data driven modeling as well as associated flood socio-economic risk map.
