This course will address selected aspects of material modeling and scientific computing in a hands-on manner. It will be composed of a lecture and a seminar part. We will discuss basic concepts like numerical issues and convergence, static vs dynamic memory allocation, object oriented programming, parallelization. The practical part will cover strategies to solve eigenvalue problems (e.g. Schrödinger’s equation), the transfer matrix method (e.g. for the reflectivity of a Bragg mirror) and Metropolis Monte Carlo methods (e.g. for magnetic domain formation), but can also adapted and extended depending on the audience. These examples will be practiced in form of homework programming assignments, which need to be solved and discussed.

Note that this is not a beginners course in programming or physics. Basic knowledge of at least one language (e.g. C/C++, Ruby, Python, …) and of common data types and structures (arrays, lists, trees, …), as well as solid-state physics is expected.

Outline:

• Basic concepts (numerical issues, static vs dynamic allocation, object oriented programming, code quality, parallelization)
• Solving eigenvalue problems using different methods
• Transfer matrix method
• Monte Carlo methods