|
In this course we want to build a parallel numerical solver for a typical eliptical partial differential equation from scratch and use it to solve in a case study a real world problem. For the sake of simplicity we use the groundwater flow equation and a contamination problem as example.
Topics of the lecture are:
- Introduction to the problem and to object oriented programming
- C++: Classes and Inheritance
- C++: Generic Programming and the STL
- Iterative linear solvers: Basics
- Iterative linear solvers: Advanced methods
- Iterative linear solvers: Multigrid
- Flow in porous media
- The Finite-Volume-Method
- Grids and grid generation
- Parallel Computers
- Parallel Programming: MPI
- Parallel Programming: Parallel iterative solvers
- Time dependend problems, Solute Transport
What you will learn: Modern programming techniques and their application to the numerical solution of partial differential equations Prerequisites: Knowledge of basic programing concepts like loops, conditions and variables. Basic knowledge about partial differential equations (content of COMMAS C5 lecture) Readiness to do some programming in the exercises.
|
JD) | © Universität Stuttgart |
Impressum