High-Performance Computing: Stuttgart's 'Hunter' Supercomputer Goes into Operation

With Hunter, HLRS is opening a new chapter in the history of supercomputing in Stuttgart. “A high-performance computer like the new Hunter is not simply a technical innovation — more importantly, it is necessary for the future capabilities of our country and a key to progress in science and industry. This in turn is a foundation for our prosperity,” says Petra Olschowski, Minister of Science, Research and Art of the State of Baden-Württemberg. “Baden-Württemberg leads Europe in the fields of supercomputing and artificial intelligence and is internationally competitive. HLRS at the University of Stuttgart plays a key role in this. Hunter will make it possible to model the next generation of safer, quieter, and more environmentally friendly aircraft like helicopters, flying taxies, and drone-supported delivery services. The additional computing performance is also relevant for progress in quantum mechanics, climatology, astrophysics, and energy research.” Prof. Peter Middendorf, Rector of the University of Stuttgart, says: “Hunter offers scientists at the University of Stuttgart and across Germany a future-proof infrastructure for AI-based simulations and high-performance computing of a new quality. Hunter also benefits the entire ecosystem of our university with its global players, its strong medium-sized companies, and its growing start-up scene.”

The HLRS system, developed and manufactured by Hewlett Packard Enterprise (HPE), is based on the latest generation of hardware and innovative software for system management. Hunter lays the foundation for the transition to the future of supercomputing at the Stuttgart site. These systems are capable of performing highly computationally intensive processes at exceptional speed, enabling more precise simulation models, the development of advanced AI algorithms, and the implementation of complex AI applications, among other uses.

Hunter is intended to help solve complex scientific problems, for example in engineering, weather and climate modeling, biomedical research and materials science. In addition, Hunter will provide public authorities and industry access to powerful, secure HPC and AI resources, especially start-ups and SMEs.

Hunter is intended to help solve complex scientific problems, for example in engineering, weather and climate modeling, biomedical research and materials science. In addition, Hunter will provide public authorities and industry access to powerful, secure HPC and AI resources, especially start-ups and SMEs.

With a theoretical peak performance of 48.1 petaflops - equivalent to 48.1 quadrillion floating-point operations per second - Hunter’s speed is nearly double that of HLRS’s previous flagship supercomputer, called Hawk. Additionally, the center has successfully implemented a new computer architecture with Hunter. In earlier systems, a large number of CPUs, i.e. central processing units, were used to increase computer performance. Hunter now uses accelerated processing units (APUs), which combine conventional processors with graphic processing units (GPUs). These graphics processors make it possible to process large amounts of data simultaneously and perform parallel calculations faster, more efficiently and much more sustainably than before. The new system architecture also includes a high-bandwidth data memory. Overall, it enables higher performance and at the same time reduces the energy consumption of Hunter at peak performance by 80 percent compared to Hawk.

Hunter is intended to support traditional high-performance computing applications, but also offers a powerful infrastructure for artificial intelligence thanks to the switch in architecture from CPUs to APU accelerators. The new system can therefore expand the HLRS user community to include data scientists and specialists who develop customized large language models and carry out deep learning projects and complex data analyses. The combination of CPUs and GPUs in a single package is designed to facilitate the development and execution of novel hybrid computing workflows that combine simulation, data analysis and artificial intelligence in innovative ways. Hunter can be used to run simulations that generate synthetic datasets for training AI algorithms, among other applications. Conversely, artificial intelligence could be integrated into simulation workflows to accelerate computationally intensive codes.

“The rapid development of AI and an increasing focus on sustainability in supercomputing mean that high-performance computing is currently going through an exciting, transformative period,” says Prof. Michael Resch, Director of HLRS. “With Hunter, our user community gains a state-of-the-art, world-class infrastructure that will support them in navigating this changing HPC landscape and enable them to remain competitive at the frontiers of scientific discovery and industrial innovation.” 

Hunter is designed as a transitional system to pave the way for HLRS's next supercomputer, called Herder, which is scheduled for installation in 2027. With a top speed of several hundred petaflops per second, Herder will constitute a major jump in peak performance over Hunter. Since Hunter has a similar GPU-accelerated architecture to Herder, the HLRS user community can already prepare their codes to take full advantage of the enormous computing power that will soon be available.

In the initial phase of the German AI factory HammerHAI, established in December 2024, Hunter will serve as the backbone for artificial intelligence operations Following its launch in 2025, HammerHAI will focus on developing solutions and services to overcome barriers to the adoption of artificial intelligence across Europe. In 2026, a new AI-optimized supercomputer of the European High-Performance Computing (EuroHPC) Joint Undertaking will be installed at HLRS, providing extensive AI computing services for German and European science and industry.

The Herder supercomputer will be housed in a new building located behind HLRS, with construction set to begin soon. HLRS III will use sustainable materials, be equipped with photovoltaic systems and use the heat generated by Herder to heat other buildings on Vaihingen campus of the University of Stuttgart. In October 2024, HLRS was honored with the Data Center Strategy Award for Transformation in recognition of its sustainable infrastructure planning.

Hunter was developed and manufactured by Hewlett Packard Enterprise (HPE). The total cost of the new system amounted to fifteen million euros. Half of the funding was provided by the Baden-Württemberg Ministry of Science, Research and the Arts (MWK) as part of its strategy for high-performance computing/data-intensive computing (HPC/DIC). The second half is provided by the Federal Ministry of Education and Research (BMBF) within the SiVeGCS project (ensuring the continued availability of GCS supercomputing resources). The funding was made possible by the Gauss Centre for Supercomputing (GCS), the association of Germany's three federal supercomputing centers.

Further Information: HLRS press information