Supercomputing

Modern science is almost inconceivable without supercomputers. The vast knowledge contained in the rapidly growing data volumes in almost all research fields can only be exploited thanks to their computing power. Supercomputers perform numerous simulations that, for example, provide insights into how potential active substances dock onto a receptor, how the climate is changing, how galaxies are formed, and how semiconductors and energy materials function.

Supercomputers of the highest performance class

Forschungszentrum Jülich is one of the leading European locations for supercomputing and provides the German and European science community with computing resources of the highest performance class. The Jülich Supercomputing Centre (JSC) offers scientists support for all aspects of supercomputing and the simulation sciences. JSC’s range of services are provided by its 200 experts and they include training courses, workshops, and conferences as well as individual consulting services covering a wide range of topics. Forschungszentrum Jülich also offers a master’s degree course and a dual study programme to train highly qualified professionals of tomorrow.

JSC does not only allocate computing time but also develops new concepts and solutions for next-generation supercomputers in cooperation with renowned manufacturers. The overall goal is to build a modular supercomputer architecture that can be used for scientific applications more flexibly and efficiently than previous systems. Such a computer will consist of several modules that can be combined with each other as required. At the same time, the concept will make it possible to integrate future technologies that are being intensively researched at Forschungszentrum Jülich, for example quantum computers and neuromorphic modules modelled on the human brain.

New challenges posed by big data and AI

The amount of data generated by science is continuously increasing. On the one hand, this is due to the fact that research can observe and measure nature ever more precisely. On the other hand, this is a result of increasingly complex simulations, like those to learn more about climate change. Machine learning and artificial intelligence are crucial tools when it comes to acquiring and processing these gigantic data sets.

This gives rise to new requirements – not only in terms of hardware and software but also for the IT infrastructure and the way in which research data are shared and collected. Forschungszentrum Jülich addresses these challenges and drives the establishment of national and European IT infrastructures for research data and standards together with its partners within the Helmholtz Association and within extensive collaborations at national and international level.

Photo copyright: Forschungszentrum Jülich / Sascha Kreklau