Welcome to the P1's HPC documentation!
The Pioneer Centre for Artificial Intelligence (P1 for short) is a Danish cross-university research centre. This documentation provides a concise overview for P1 affiliates on using our own HPC facilities and relevant partners.
| System | Tier | Price (compute) | Price (storage) | GDPR data |
|---|---|---|---|---|
| P1 DTU | 3 | Free to P1 users | Free to P1 users | No |
| P1 NGC | 3 | Free to P1 users | Free to P1 users | Yes |
| Gefion (Partner) | 1/2 | Pay-per-hour | Pay-per-month | No (not yet) |
For additional details on P1, visit the P1 Homepage and P1 Computing.
The goal of the P1 HPCs is to enable research opportunities for affiliates of P1, supporting projects of various scales. While we prioritize smaller projects as they represent the majority of research needs, we also support larger projects through our tiered compute resources, including the capability to train very large models and support some European-level workloads on Tier 2/1 facilities.
However, since resources are limited and we aim to serve as many P1 members as possible, very large scale projects might be better supported within their own funding. The compute clusters are primarily intended for research projects. While commercial use is generally not permitted on the P1 DTU and NGC HPCs, the Gefion HPC may accommodate certain commercial research projects on a case-by-case basis, subject to approval and specific terms.
In this documentation, we cover only a broad overview of the available HPC facilities and basics for getting started. Since the clusters are installed at separate institutions, the documentation on the technical side of things differ by facility:
When having used support from DCC (the DTU HPC center administrating our DTU cluster) please cite the following reference: