AI Systems
AI Systems / BayernKI
Click for Details
The LRZ AI Systems are designed to support the scientific community in the fields of Big Data and Artificial Intelligence, with a particular emphasis on GPU-intensive workloads. The compute resources are partially funded through the BayernKI initiative.
Overview
1. Access - Valid LRZ user account and AI-enabled project required for access.
2. Compute - Available GPU/CPU partitions and hardware (P100, V100, H100, etc.).
3. Storage - Explanation of the DSS storage tiers (home, AI Systems DSS, Linux Cluster DSS).
4. Enroot - Introduction to the Enroot container runtime for unprivileged, high-performance images.
5. Slurm - Guidance on using Slurm for interactive and batch jobs (single GPU, multi-GPU, multi-node).
- 6. Interactive Apps - Run Jupyter, RStudio, and TensorBoard directly through the web interface.
7. Datasets and Containers - Repository of shared public datasets and Enroot images.
8. Partners - Lists partner organizations and collaboration context.
9. Changelog - Historical record of system updates, maintenance events, and version changes.
Regulations, Terms and Conditions
Users of the AI Systems need to adhere to the following regulations, terms and conditions:
- General usage regulations (nonbinding translation) / Benutzungsrichtlinien.
- Access to the AI Systems is subject to export control regulations.
- LRZ stores and publishes your project data and description electronically. This requirement can be waived upon explicit request.
- A report on the work conducted and results obtained by the use of the AI Systems must be provided upon LRZ request.
Acknowledgement
Please use the following template for the acknowledgement of resources and support provided by the LRZ.
Next - 1. Access
