Target group
- Scientific research, companies, public institutions
Your requirements
- Use of high-performance computing (HPC) resources
- Need for GPU-based systems for inference tasks
- Low-barrier access
- Highly available inference platform
- Guaranteed time-to-completion
Our offer
For the applications and services of the AI service centre, a GPU-based HPC system with current NVIDIA H100 GPUs is provided for inference tasks in research, development and technology.
The computing nodes are connected to each other and to the storage resources also provided via a RoCE network.
The inference platform is installed geo-redundantly at the sites of the Leibniz Universität IT Services (LUIS) in Hannover and the Gesellschaft für wissenschaftliche Datenverarbeitung mbH Göttingen (GWDG).
Hardware-wise it features 9 and 12 GPU nodes, respectively, each equipped with 4 NVIDIA H100 PCIe GPUs (80 GB HBM2e memory).
At the software layer operations are realized with Slurm and Kubernetes. The storage is synchronized between both sites.
Software, models and data can be installed via self-service or integrated via the KISSKI catalogues.
The system can either be used by direct access or as a technical basis for the KISSKI services. prerequisites: | For direct use of the inference platform, a current SSH client is required.
Individual requirements apply for indirect use of the inference platform through the services offered by KISSKI.
New services can be setup based on Kubernetes.
