Mario Wille, M.Sc.

Technische Universität München

TUM School of CIT
Department of Computer Science
Boltzmannstrasse 3
85748 Garching
Germany

E-Mail: mario.wille(at)tum.de 
Tel: +49-89-35831-7816 (usually easier to reach me via email)
Office: at Leibniz Supercomputing Centre (Boltzmannstr. 1), E.2.048
Office Hours: by arrangement
GitHub, ORCIDResearchGateGoogle Scholar

Research Interests

  • High Performance Computing
  • Parallel Algorithms
  • Parallelization of HPC Applications
  • Extreme-Scale Simulations
  • Computational Seismology
  • GPU Programming

Teaching

Winter term 2023/24

Course no.TitleLecturer (assistant)DurationTypeDates
0000000791Bachelor-Practical Course Tsunami-Simulation (IN0012, IN4153)
  • M. G. Bader [L]
  • M. Marot-Lassauzaie
  • M. Wille
6PRLink
0000001052Turbulent Flow Simulation on HPC-Systems (IN2311, MW2048)
  • H.‑J. Bungartz [L]
  • S. Narvaez Rivas
  • M. Wille
3VILink

Open Student Projects

I mainly work with the ExaHyPE Engine.

If you're interested in a Master's or Bachelor's thesis write me an E-Mail with your interests and preknowledge.

Topics are normally placed around my research interests and our current research projects:

  • Tsunami Simulation
  • Computational Seismology
  • Hyperbolic PDEs
  • Efficient GPU Offloading

For more information, you can also have a look at the software projects I am involved with.

Running Student Projects

Jörn Fischbach: Load Balancing on Dynamically Adaptive Grids. Bachelor thesis, 2023

Siba Zgheib: Multi-GPU Implementation of a Hyperbolic Finite Volume Solver in ExaHyPE. Master thesis, 2023

Finished Student Projects

2023

2021

Publications

2023

  • Mario Wille; Tobias Weinzierl; Gonzalo Brito Gadeschi; Michael Bader: Efficient GPU Offloading with OpenMP for a Hyperbolic Finite Volume Solver on Dynamically Adaptive Meshes – HPC Algorithms and Applications. In: High Performance Computing. Springer (13948. Aufl.), 2023, 65-85 mehr… BibTeX Volltext ( DOI ) Volltext (mediaTUM)

2022

  • Yakup Budanaz; Mario Wille; Michael Bader: Asynchronous Workload Balancing through Persistent Work-Stealing and Offloading for a Distributed Actor Model Library. 2022 IEEE/ACM Parallel Applications Workshop: Alternatives To MPI+X (PAW-ATM), 2022SC22 Workshops mehr… BibTeX Volltext ( DOI ) Volltext (mediaTUM)

Talks

2023

  • Mario Wille: Efficient GPU Offloading with OpenMP for a Hyperbolic Finite Volume Solver on Dynamically Adaptive Meshes. High Performance Computing, Springer, 2023 38th International Conference, ISC High Performance 2023, Hamburg, Germany, May 21–25, 2023, Proceedings mehr… BibTeX Volltext (mediaTUM)

2022

  • Mario Wille: Asynchronous Workload Balancing through Persistent Work-Stealing and Offloading for a Distributed Actor Model Library. 2022 IEEE/ACM Parallel Applications Workshop: Alternatives To MPI+X (PAW-ATM), 2022SC22 Workshops mehr… BibTeX Volltext (mediaTUM)

Posters