JUPITER highlights exascale science at ISC

JUPITER, Europe’s first exascale supercomputer at Germany’s Forschungszentrum Jülich, is running major science and AI workloads on NVIDIA Grace Hopper Superchips and NVIDIA Quantum-X800 InfiniBand networking. At ISC in Hamburg, projects on the system span brain mapping, climate modeling, wireless network research and quantum computing.

The Jülich Brain Atlas effort has produced CytoNet, a foundation model for brain microarchitecture analysis. Training ran on JUPITER in under five days, using 6.5 petabytes of data from 21 post-mortem brains on 4,096 NVIDIA Grace Hopper Superchips. The team’s next step is an AI agent for brain researchers that combines multimodal reasoning, language interfaces and Q&A capabilities using open models, including NVIDIA Nemotron 3 120B.

Climate researchers are using ICON to simulate a coupled Earth system at 1-kilometer resolution, covering ocean, atmosphere, land, biogeochemistry and the full carbon cycle. Running on 20,480 NVIDIA Grace Hopper Superchips, the model simulated roughly 146 days of real climate into 24 hours of compute, setting a world record in global climate simulation.

Ericsson and Forschungszentrum Jülich are also using JUPITER to develop AI for the continued evolution of 5G and for 6G networks, with a focus on energy-efficient inference and brain-inspired architectures. Jülich researchers and the NVIDIA Application Lab also fully simulated a universal 50-qubit quantum computer, surpassing the previous 48-qubit record and creating the JUQCS-50 simulator for quantum algorithm research.