Europe's First Exascale Supercomputer Jupiter Powers Science

If you want to visualize the surface air flowing across every meter of the planet Earth, you need to plot nearly three trillion squares of land and sea and space. To do that you need a mind-bogglingly big calculator.

Which is why Ioan Hadade, a computational scientist working with vast weather forecasting and climate models, is excited about the machine now online an hour down the road from his lab in Bonn, Germany. Europe’s first exascale supercomputer—called JUPITER, after a much bigger planet than our own—is nearly fully operational. It is currently running scientific programs on its formidable processors.

JUPITER debuted at No. 4 in the June 2025 global TOP500 list of the world’s most powerful computer systems.

It is based at the Jülich Supercomputing Center in the German Rhineland between Cologne and Aachen, running on a booster module with 5,900 accelerating compute nodes. Some 24,000-odd Nvidia Grace-Hopper superchips give JUPITER its oomph; the machine also features a universal cluster module with 1,300 nodes using Rhea1 processors, and an InfiniBand NDR network for the high-speed interconnects.

The semi-annual TOP500 rankings are a way to engage every single element of a machine for performance. Benchmarking proves the functionality of a highly complex operation. “And now, it’s better to have some science done on the machine,” says Thomas Lippert, director of the Jülich Supercomputing Center.

Computational Science at Scale

As of mid-June research enterprises were on the JUPITER machine testing scientific calculations. “You need a really large machine to run this,” Hadade says. He’s referring to the Destination Earth digital twin projects he and his colleagues are part of developing at the European Center for Medium-Range Weather Forecasts.

The Destination Earth digital twins are replicas of Earth systems used to monitor, simulate, and predict the interaction between natural phenomena and human activities. Hadade and his colleagues have produced physics-based observations of atmospheric conditions and weather at 9-kilometer resolutions and closer. Being able to zoom in—to show physical phenomena at a resolution of 700 meters or finer—reveals the processes that prompt deep convection and turbulence. The team is turning to JUPITER to do it.

Across the country at the Technische Universität Ilmenau, physicist Jörg Schumacher is investigating convection and turbulence by visualizing the intricate flow of thermal plumes, the kind you might find within cloud formations or, in a more violent state, on the surface of the sun. Convection is driven in the simplest case by the flow of fluids or gases through different temperature bands: You could have a hot bottom and cold top layer, and if the temperature difference is strong enough, it starts to get very turbulent between them.

Highly nonlinear filaments, vortices, and eddies start forming patterns and structured networks. “Everything should be very chaotic, irregular, stochastic—but it isn’t,” Schumacher says. “How is nature forming these nice beautiful patterns in highly turbulent flows?”

A supercomputer called JUWELS produced this astoundingly detailed image of temperature variations in a thermal plume. JUPITER will be able to zoom in even closer.Jörg Schumacher/Technische Universität Ilmenau

Schumacher and his colleagues have run these visualizations on supercomputers allowing for astounding detail. They’ve done this work on JUWELS, a Jülich supercomputing booster module; they’ve reached a point where they cannot do larger jobs on it. They’ll use JUPITER to explore further.

Hadade says more than rankings, what really matters is having ready access to these machines. “It shows a commitment that Europe is really taking high-performance computing seriously,” he says.

The Energy of Supercomputing

The exascale planning for JUPITER began in earnest in 2018 with the founding of the European High-Performance Computing Joint Undertaking (EuroHPC JU). It developed through the pandemic years: In 2024, the JEDI prototype and JETI transition system module came online, and JUPITER was fully installed by the end of April this year. European plans are in swing for a second exascale system, a €544 million machine to be called Alice Recoque and hosted in Bruyères-le-Châtel, south of Paris.

These ambitions are not just about size and speed but also reflect efficiency and sustainability concerns for energy consumption and cooling. JUPITER’s cooling system circulates water from the nearby Rur river and will use excess heat to warm campus buildings during colder months.

JEDI, the JUPITER prototype, leads another list, the GREEN500; JUPITER itself premieres at No. 21. Lippert says the difference there is scale: Moving to a larger machine means parallelization losses. There is also physics in play. Lippert expects efficiency improvements as JUPITER is optimized; it is already the most energy efficient of the five fastest systems in the world, with more than 60 billion floating-point operations per watt.

Even so, at its highest load JUPITER consumes about 17 megawatts of energy. As more and more massive data centers come online to meet demands for artificial intelligence and technology-saturated daily living, out-of-control energy usage is a pressing anxiety.

Researchers like Schumacher and Hadade already see enough convection and temperature anomalies in their visualizations. “Anything that makes the machines have less of an impact on the environment is welcome,” Hadade says. “We need to be careful that we don’t need a nuclear power plant next door each time to power these things.”