Master Thesis MSTR-2022-35

BibliographyKünzel, Sebastian: Design and implementation of a GPU-based simulation- and analysis-environment for dynamical systems.
University of Stuttgart, Faculty of Computer Science, Electrical Engineering, and Information Technology, Master Thesis No. 35 (2022).
107 pages, english.

In this thesis, a high-performance GPU-based simulation- and analysis-environment for dynamical systems is designed, implemented, and tested. Computation is optimized on the basis of the GPU hardware architecture. Minimal task-specific simulation programs are created using runtime compilation. Scan execution and analysis methods are designed and optimized for massive parallelization and integrated into a modular, efficient, scalable, and streamlined architecture. The new program is tested for performance and result accuracy. The impact of floating-point inaccuracy is discussed and shows how the current implementation of AnTGPU can mitigate this problem.

Full text and
other links
Department(s)University of Stuttgart, Institute of Parallel and Distributed Systems, Applications of Parallel and Distributed Systems
Superviser(s)Schwarz, Prof. Holger; Avrutin, Prof. Viktor
Entry dateSeptember 16, 2022
   Publ. Computer Science