The development of hardware is progressing towards hybrid architectures with CPU / GPU components or mainly GPU components, which needs an experienced user and developer community with a knowledge of these more complex hardware architectures. Since years Juelich fosters the area of "Modular Supercomputing", which aims at coupling a variety of hardware architectures for scientific applications, so that problem specific hardware components can be easier chosen than within a monolithic organisation of hardware. This concept will be further considered and developed in future. Therefore, the students first get an intense introduction to different programming concepts. A selection from a variety of research topics will be considered individually for each student and will be balanced with the current research activities of the scientific groups at Forschungszentrum Jülich, where the students will be hosted. Students can express their interest in a given research topic already in their application letter and scientific supervisors are selected in order to optimise the match between the student's interest and the current research activities at the institute.

The school focuses on, but is not limited to, the following domains and their application to HPC:
- Parallel algorithms
- Performance Analysis
- Molecular dynamics
- Monte Carlo
- Electronic structure / DFT
- Linear algebra
- Biophysics
- Statistical Physics
- Quantum Computing
- Data Science
- Machine Learning
- Weather/climate modelling
- Quantum chemistry
- Visualisation