The High-Performance Computing Center Stuttgart (HLRS) of the University of Stuttgart is one of the three member centres of the Gauss Centre for Supercomputing (GCS). HLRS supports national and European researchers from science and industry by providing high-performance computing platforms and technologies, services and support. Services are provided in cooperation with the other two member centres of GCS and with industrial partners such as hww GmbH (T- Systems, Porsche AG) and SICOS BW GmbH.
HLRS conducts basic and applied research in the field of high-performance computing in publicly funded national and
European projects as well as industrial research activities in association with partners from research and industry.
Collaborative research in the automotive field is carried out in cooperation with the Automotive Simulation Center
Uncertainty Quantification (UQ) is widely used in quality control, risk management and engineering simulation. Monte Carlo method is one of the most popular and intuitive method to perform a UQ analysis. But it is normally very time consuming to apply a Monte Carlo method to a large CFD problem, since we need to run the CFD code repeatedly until it
converges. Now we are striving to find out a way to transform the code in a UQ intrinsic code, so that we can analysis how is the uncertain transported throughout the code and achieve better performance by vectorization etc. In order to achieve this goal, we plan to establish a schematic way to analysis to the code, so that we can figure out how the uncertainty transported though the code. Furthermore, it is worthy to develop an automatic analyzing tool to perform such process.
Your Task (the following choices can be discussed according to your interest):
- Use python to write a parser and a Lexer liked analyzer, which can detect the places, where the uncertainty is
- Apply the analysis method to the code, vectorizes the code and test its performance, compare with the existed
Monte Carlo method. Record the result of your work
What you will get:
1. Programing experience with python and Fortran
2. Learn the basic concept of compiling, how to build a parse and a lexer
3. Familiar with basic concept of parallel computing
4. Learn how to analysis your code and optimize your code
5. Flexible working hours and relaxable working environment (as soon as you can work on site)
We hope you bring:
- You are student of Computer Science, Aerospace Engineering, or related course of study
- Familiar with programming, if you interested with the first task. Familiar with python and basic knowledge
with Fortran. It is a plus if you already have knowledge with working mechanism of Compiler.
- Experience with Linux OS
- Motivated and creative thinking
- Working at least 40-60h Pro Month, it will be a plus if you can work long-term
If you are interested, please contact: firstname.lastname@example.org