Brain and Computing Power

Many brain researchers use the Jülich supercomputers. Sandra Diaz (41), Wouter Klijn (44), and their colleagues from JSC are on hand to help.

Sandra Diaz (41) und Wouter Klijn (44) vom JCS — Copyrights: Limbach /Forschungszentrum Jülich

What exactly do you do to support brain researchers?

Diaz: Briefly speaking, we use our technical expertise to help them get the most out of our powerful supercomputers. This allows them to work better and much faster, and sometimes even discover completely new possibilities. We show them various technical options including their advantages and disadvantages, and also help them to write good and successful applications for valuable computing time.

Klijn: Developing a process for typical applications in brain research, known as use cases, together with the scientists was very helpful for this collaboration. This often allows us to identify existing tools that are already available for such research questions or to adapt software with similar needs to the current case with relative ease.

How do you ensure that research and the computing world understand each other?

Klijn: We first had to familiarize ourselves with the work of brain researchers and we need to stay up to date with developments to understand what they do and what they need. We are now able to communicate in such a way that the researchers understand us. It’s not a case of us simply handing them the keys to the “Ferrari”, but making sure that they can also drive it to the finish line. It took several years of discussion to reach this point.

Where does supercomputing help brain researchers the most?

Diaz: On the one hand, there are the huge amounts of data that need to be processed, visualized, and analysed. On the other hand, supercomputing is needed to carry out simulations of brain functions as quickly as possible. For individualized clinical applications, “digital twins” of the brains of individual patients will be created on the computer in future and linked with the numerous and complex data from EBRAINS. This would be unthinkable without supercomputing and its parallel computing.

What do you mean by parallel computing?

Klijn: A normal computer is like a single employee who takes on a big task alone. A supercomputer is like a whole team of employees all working on the same big problem at the same time. Of course, this makes everything much more complicated, but the calculations are also much faster.

How will the next generation of brain researchers learn to deal with all these new digital possibilities?

Diaz: In the German National Node of EBRAINS, we offer training courses and workshops to prepare neuroscientists for the use of supercomputing. Our next step is to integrate computing skills and EBRAINS tools into neuroscience degree courses at German universities.

Originally published in the employee magazine "intern" of Forschungszentrum Jülich.
Author: Hanno Schiffer