ChEESE

The Center of Excellence for Exascale in Solid Earth (ChEESE; https://cheese-coe.eu) will address scientific and societal challenges by harnessing European institutions in charge of operational monitoring networks, tier-0 supercomputing centers, academia, hardware developers and third-parties from SMEs, industry and public-governance. The scientifically challenging ambition is to prepare 10 open-source flagship codes to solve Exascale problems on computational seismology, magnetohydrodynamics, physical volcanology, tsunamis, and data analysis and predictive techniques from monitoring earthquake and volcanic activity. The codes will be optimized at both intranode level and internode level on heterogeneous hardware prototypes for the upcoming Exascale architectures, thereby ensuring commitment with a co-design approach. Preparation for Exascale will also consider aspects of scientific simulation workflows such as data management and sharing, I/O, post-processing and visualization.

The ChEESE Center of Excellence builds on three pillars: pilot demonstrators, services and training. First, the project will develop pilot demonstrators for scientific challenging problems requiring of Exascale computing in alignment with the vision of European Exascale roadmaps. This includes near real-time seismic simulations and full-wave inversion, ensemble-based volcanic ash dispersal forecasts, faster than real-time tsunami simulations and physics-based hazard assessments for seismics, volcanoes and tsunamis. Second, pilots are also intended for enabling of operational services requiring of extreme HPC on urgent computing, early warning forecast of geohazards, hazard assessment and data analytics. Selected pilots will be tested in an operational environment to make them available to a broader user community. Finally, the third pillar of ChEESE aims at acting as a hub to foster HPC usage across the Solid Earth Community and related stakeholders and to provide specialized training on services and capacity building measures.

HLRS mainly contributes to the performance analysis and optimization of the flagship codes, in particular in relation to inter-node scalability and I/O aspects. Further, HLRS leads the effort to prepare the scientific workflows of the pilot demonstrators for future Exascale environments.