Thesis / Interdisciplinary Project (IDP) / Research Practice / Study Project
30.01.2025, Abschlussarbeiten, Bachelor- und Masterarbeiten
to assist with the Containerization of (Computational Fluid Dynamics) Workflows on High-performance Computing Systems (HPC)
for TUM-students within Informatics, Aerospace, Mechanical Engineering, Data Science or similar
Part of good scientific practice, is ensuring reproducibility of conducted research. In this work, the workflow of a CFD simulation from compilation over parallel execution to post-processing is moved into a container-based equivalent. The containered setup is analyzed for practicability and performance on the compute resources pro-vided by the Leibniz Supercomputing Centre (LRZ), Jülich Supercomputing Centre (JSC) and High-Performance Computing Center Stuttgart (HLRS).
Tasks
• Familiarization with HPC workflows
• Familiarization with JUWELS supercomputer (possibly also HLRS)
• Familiarization with container software (Apptainer)
• Container setup and performance improvement
• Documentation
Requirements
• Knowledge of Python or C++
• Knowledge of Linux-CL
• Knowledge of GPU computing
• Knowledge of containers (e.g. Docker or Apptainer)
• Experience with HPC-clusters (beneficial)
Links
• GPU Computing at JSC: https://apps.fz-juelich.de/jsc/hps/juwels/gpu-computing.html
• Container Runtime at JSC: https://apps.fz-juelich.de/jsc/hps/juwels/container-runtime.html
• GitLab “Containerization”: https://gitlab.lrz.de/nfdi4ing/containerization
• NFDI4Ing research group: https://www.epc.ed.tum.de/en/aer/research-groups/nfdi4ing/
Contact
Benjamin Farnbacher
benjamin.farnbacher@tum.de
089.289.16094
Kontakt: benjamin.farnbacher@tum.de
Mehr Information
Announcement_as_PDF |
Announcement_as_PDF,
(Type: application/pdf,
Größe: 161.0 kB)
Datei speichern
|