Index

Since 2022, NHR@FAU has its GPU cluster “Alex” in operation, which contains a total of 192 NVIDIA A100 and 304 NVIDIA A40 GPGPUs with eight cards per node. To allow for an easier comparison with the already installed GPUs in the “TinyGPU” cluster (NVIDIA RTX2080Ti, NVIDIA RTX3080, and NVIDIA V100) a...

The ISC 2025 in Hamburg is one of the best places in the world to be for learning about high-performance computing and its applications.  Its goal is to spread knowledge, foster innovation, and promote international teamwork in this area through its rich program of keynotes, conference talks, worksh...

The supercomputer "Helma" has been extended in April 2025 with another 384 NVIDIA H200 GPUs. Comprising 768 NVIDIA GPUs (H100/H200), it is now the most powerful system at German universities. Bavaria's Science Minister Markus Blume has officially started the early user access phase on the full syste...

Since about 6:00 p.m. today, the parallel filesystem ($FASTTMP) is partially down. Jobs on Fritz, Alex, and Helma may be affected. We are working on solving the issue and apologize for the incovenience. Update : Lustre is fully available again.  

The Fritz and Helma clusters are down today (May 14) because of urgent maintenance of the water cooling system. We apologize for the inconvenience and will keep you updated about the status. Update About half of the Fritz cluster nodes are up and running again. Update The Fritz and Helma clu...

The next HPC Café will take place on Tuesday, May 13, 2025, at 4:00 p.m. as a hybrid event. As always, there will be plenty of time to get in touch with your favorite HPC group. We invite you to come to NHR@FAU to enjoy coffee, cake, and computing. The event starts at 4:00 p.m. CEST with an open ...

On April 29 and 30, the SCS Hackathon 2025 took place at FAU, jointly organized by the NHR@FAU and the Sovereign Cloud Stack project. 22 participants from various places, even from the Netherlands and Slovakia, were welcomed to share creative ideas and impressive solutions. For two days, members ...

There will be scheduled downtime of our Tier-3-clusters TinyGPU and TinyFat, including the frontend tinyx and special machines like memoryhog, starting on Wednesday, May 7 at 09:00 and lasting at least the whole day. Reason for the downtime is an OS update. Update 17:00: the frontend tin...

We had to perform an emergency shutdown for the clusters Fritz and Helma today (Wednesday, April 16) at around 13:30, after the cold water supply for these clusters failed at around 13:10. All jobs running at that time were terminated (ungently). Update 16.04. 15:50: First analysis showed that a ...

On Apr 8, 2025, all NHR Graduate School scholars in Erlangen participated in an in-person joint seminar with the local liaison scientists team. The goal was to get to know each other and to foster interaction and collaboration. In his introductory note, Prof. Gerhard Wellein, head of NHR@FAU, emphas...