Shallow water simulations are important for climate models, flood or tsunami predictions and other applications. Performing such simulations on unstructured meshes with the Discontinuous Galerkin method is numerically attractive, but a performance challenge on conventional architectures. With a customized dataflow architecture implemented on FPGAs, we have improved performance and power efficiency on a single FPGA and achieved promising initial results when scaling to multiple FPGAs via direct FPGA-to-FPGA interconnects.

Over the past decades, large-scale computer simulations have grown to become one of the most powerful approaches to study the interior of Earth-like planets. Geodynamical models are used to investigate the evolution and distribution of the temperature inside the planet that ultimately affects its structure and the way the planet cools over time. Combined with data obtained from planetary missions and laboratory experiments, these models help us to improve our understanding of the history and current state of planets in our Solar System and beyond. These models can teach us about the formation and evolution of planetary environments

ICON-ART is the next generation model for seamless simulation of numerical weather forecast, climate prediction and atmospheric composition modelling. It is a joint development project of DWD (German Weather Service), MPI-M (Max-Planck-Institute of Meteorology), DKRZ (German Climate Computing Center), and KIT (Karlsruhe Institute of Technology). At IMK-ASF the simulations focus on the understanding of composition-climate interactions as well as atmospheric chemical and microphysical processes on different scales

PALM-4U is a newly developed high-resolution urban-climate model. It is designed as a tool for researchers and city planners to simulate and analyze the urban climate and its effects on city dwellers. The key feature of PALM-4U is its capacity to directly resolve turbulence effects and provide highly accurate simulation results. Apart from that, PALM-4U offers further features such as sophisticated bio-climate and air chemistry analysis or a multi-agent model that simulates individual city dwellers wandering across the city. To gain confidence in PALM-4U, extensive evaluation is necessary.