NHR4CES@TUDa
Technische Universität Darmstadt
Darmstadt
NHR4CES@TUDa, in collaboration with NHR4CES@RWHT, focuses on providing hardware and scientific support, particularly for engineering applications and their interface with materials and the life sciences.
Key Expertise and Profile
- Flagship Collaboration: NHR4CES@RWTH and NHR4CES@TUDa bring together two NHR centers to combine their expertise and create strong synergies
- NHR4CES offer cutting-edge computing infrastructure with our state-of-the-art. Beyond outstanding hardware, dedicated experts offer scientific support and a wide range of training opportunities
- Cross-Sectional Groups (CSGs) and Simulation and Data Labs (SDLs) support researchers in using HPC innovatively
- The CSGs at NHR4CES@TUDa has advanced key methods relevant across HPC: such as data engineering and AI, as well as parallelism and performance
- The SDLs connect this expertise to specific fields such as Energy Conversion, Fluids, and Materials Design, bridging infrastructure and application for impactful research.
NHR4CES@TUDa is also a member of HKHLR, the regional HPC competence network of Hessia.
Technical Specifications
| Center Name | NHR4CES@TUDa |
|---|---|
| Total Resources: | |
| # CPU-Cores | 120 888 |
| # GPUs | 100 |
| Additional Information | |
| max. job run time [h] | 168 |
| jupyter notebook available [y/n] | n |
| CPU Resources: | |
| System 1 | Lichtenberg II Stage 1 |
| CPU-Type | Intel Cascadelake |
| # nodes | 632 |
| # Cores per node | 96 |
| memory per node [GB] | 384 - 1536 |
| System 2 | Lichtenberg II Stage 2 |
| CPU-Type | Intel Sapphire Rapids |
| # nodes | 579 |
| # Cores per node | 104 |
| memory per node [GB] | 512 - 2048 |
| GPU Resources | |
| System 1 | Lichtenberg II Stage 1 |
| GPUs per node | 4 x Nvidia V100 |
| # nodes | 4 |
| memory per GPU [GB] | 32 |
| System 2 | Lichtenberg II Stage 1 |
| GPUs per node | 4 x Nvidia A100 |
| # nodes | 7 |
| memory per GPU [GB] | 40 |
| System 3 | Lichtenberg II Stage 2 |
| GPUs per node | 4 x Nvidia H100 |
| # nodes | 2 |
| memory per GPU [GB] | 80 |
| System 4 | Lichtenberg II Stage 2 |
| GPUs per node | 4 x Intel Ponte Vecchio |
| # nodes | 5 |
| memory per GPU [GB] | 128 |
| System 5 | Lichtenberg II Stage 2 |
| GPUs per node | 8xAMD MI300x |
| # nodes | 2 |
| memory per GPU [GB] | 192 |