Loading
At a glance | Challenges | Solutions | Contact | Teams & Labs | Research Groups
At a glance
The aim of the topic “Supercomputing & Big Data Infrastructures” is to develop, procure, and operate the most powerful data analysis and simulation infrastructures, and to make them accessible to a wide range of users. The Jülich supercomputers JUWELS and JURECA currently rank among the most powerful in the world. Jülich will also be home to Europe’s first exascale supercomputer, JUPITER.
Jülich researchers develop both software and hardware for these powerful machines, as well as innovative modular supercomputer architectures. These architectures also permit the integration of quantum computers and future neuromorphic systems. An additional focus of this topic is quantum technology, which Jülich scientists investigate from basic research to application.
Challenges
Extensive computer simulations, large-scale data analyses, machine learning, and artificial intelligence require an enormously powerful IT infrastructure. JUPITER is set to become the first supercomputer in Europe capable of surpassing the threshold of one quintillion floating-point operations per second, taking scientific simulations to a new level and enabling breakthroughs in the use of artificial intelligence.
At the same time, new technologies of the future such as quantum computers and neuromorphic computing are opening up completely new avenues for processing and interpreting data.
Solutions
Jülich researchers operate two of the world’s current fastest supercomputers, JUWELS and JURECA, and support other researchers in using them. The performance of the current supercomputers will be topped by JUPITER, which is currently being built on the Jülich campus. JUPITER will be Europe’s first exascale computer. It will be capable of performing more than one quintillion floating-point operations per second, and even more for AI workloads. This makes it an outstanding tool for climate research, the search for new types of drug, the development of renewable energies, or the training of large AI models, for example.
Jülich scientists develop AI foundation models for numerous applications in science, including for climate, brain, and materials research. Jülich is also developing large image and language AI models such as OpenGPT-X, which comply with European data protection regulations.
Experts at the Jülich Supercomputing Centre (JSC) also collaborate with national and international partners to test and develop new concepts and technologies for next-generation supercomputers. These supercomputers will integrate various technologies that can be combined in different ways depending on the requirements – thanks to a dynamic modular supercomputing architecture developed at Jülich in cooperation with external partners from science and industry. In future, digital supercomputers will be connected with quantum computers and neuromorphic systems as part of a modular design, bringing the vision of a hybrid supercomputer within reach. New types of computers will thus be created that can be used flexibly for different research approaches.
To achieve this goal, Jülich experts are working intensively on the basic research and possible construction of quantum computers. Calculations show that these computers will be able to solve tasks in minutes that would currently take the world’s fastest supercomputer several thousand years. JUNIQ, the Jülich UNified Infrastructure for Quantum computing, already provides science and industry with access to various experimental systems, prototypes, and commercially available instruments in the quantum cosmos. It also supports users in the development of algorithms and applications for quantum computing.
Contact
Jülich Contact Person
Prof. Dr. Estela Suarez
Joint Lead of JSC-Division "Novel System Architecture Design" Spokesperson of Helmholtz Information Program 1, Topic 2, and PI in Topics 1 and 2
- Institute for Advanced Simulation (IAS)
- Jülich Supercomputing Centre (JSC)
Building 16.4 /
Room 222
Room 222
+49 2461/61-9110
E-Mail
Principal Investigators
teams and labs
Algorithms, Tools and Methods Labs
Research Groups
No results found.
Loading
Last Modified: 08.03.2025