Welcome to P1's HPC Documentation
Your overview guide to high-performance computing resources at the Pioneer Centre for AI in Denmark
General Purpose HPC
The P1 DTU HPC cluster is designed for general research workloads. It provides free compute and storage resources for P1 affiliated researchers, making it ideal for a wide range of academic and scientific projects. This cluster is not GDPR-compliant and is best suited for non-sensitive data and standard research tasks.
Sensitive HPC
The P1 NGC HPC cluster provides a GDPR-compliant computing environment with free compute and storage resources for P1 affiliated researchers. It is specifically designed for projects that require handling of sensitive or personal data, ensuring compliance with data protection regulations while supporting advanced research needs.
Large Scale HPC
The Gefion HPC is a partner facility offering pay-per-use compute and storage resources for larger scale projects. It is ideal for users with high computational demands that exceed the capacity of the standard P1 clusters. Access is subject to approval and specific terms.
-
Intended Use
Our compute clusters are primarily intended to support research projects. Commercial use is generally not permitted on the P1 DTU and NGC HPCs; however, Gefion may accommodate certain commercial research projects on a case-by-case basis.
-
Resource Policy
Resources are limited and are primarily prioritized for P1 members at the PhD, postdoc, and faculty levels. We are committed to open access and strive to serve as many P1 members as possible. For very large-scale projects, we encourage you to reach out to discuss potential solutions.
-
Detailed Cluster Information
Each cluster is hosted at a separate institution and managed by its own administrators. This site provides an overview and a starting point for accessing P1 compute resources. For the most up-to-date technical details, please refer to the respective cluster websites and support channels.
-
Citation
If you have received support from DCC (the DTU HPC center administering our DTU cluster), please cite: http://doi.org/10.48714/DTU.HPC.0001. A great deal of effort goes into ensuring that compute resources are available and that the infrastructure runs smoothly. Your citation helps recognize and support this work.