International Guest Student Programme on Scientific Computing
The International Guest Student Programme on Scientific Computing offers young students the possibility to learn and practice concepts from parallel computing. They are integrated into scientific groups and supervised by experienced researchers with a solid background in scientific computing and experience in parallel computing, in fields as soft matter, plasma physics, statistical physics, biophysics, computational fluid dynamics, materials science.
The programme starts with an introductory course about basic techniques of parallel computing as well as using Juelich supercomputers (JURECA – GPU/KNL Intel cluster). The course consists of lectures and practical hands-on sessions. Each student is then assigned to a scientific core group where he is working on a project within the context of ongoing research interests of this group. In the end of the programme, a 2-day seminar is organized where the students give a 30 minutes talk about their project during their stay in Jülich. In addition each student prepares a final report (~10 pages each) which is published as techincal report.During the programme, the students are employed as student assistants (19 hours/week).
The participants are expected to be resident in Jülich for the entire period. Forschungszentrum Jülich will provide assistance in finding accomodation in Jülich.
Week 1: Introduction to parallel computing
- Introduction to the Jülich compute facilities
- Message passing interface (MPI)
- Basics of GPU computing
Week 2-10: work on individual research projects in scientific groups, e.g.
- Parallel algorithms
- Molecular dynamics
- Monte Carlo
- Electronic structure / DFT
- Linear algebra
- Plasma physics
- Weather/climate modelling
- Quantum chemistry
Week 9-10: preparation of the final report of the individual research project
Week 9: oral presentation of research results from individual projects
Ivo Kabadshow (Research Centre Jülich) - Organiser
Godehard Sutmann (Forschungszentrum Juelich) - Organiser