NEMO2 Production Mode

Published: 14 May 2025 by NEMO Team

NEMO2 started slowly in January with only a few Genoa nodes and some test users until we went into production mode with Genoa nodes in April. In May, after the end of NEMO1, we also added most of the Milan nodes to NEMO2. This week we added our AMD Instinct MI300A APU and Nvidia L40A GPU nodes.

NEMO1 has reached the end of its life. However, there are still 25 Milan nodes and one Nvidia V100 node available until the end of May when we officially shut down NEMO1 (31.05.2025). After that, we will try to keep BeeGFS online for data transfers until we know when the new AI nodes will be delivered. But we will then shut down BeeGFS with only a few days notice. Please try to migrate your data by May 31st. After this date, we can no longer guarantee data availability.

If you still want or need to perform calculations on NEMO1 until May 31st, please write a ticket with your NEMO1 username. We will then activate your user for NEMO1.

For NEMO hardware and Slurm partitions visit our wiki.

NEMO2 has officially launched, transitioning from a testing phase to full production with expanded hardware, including AMD Instinct MI300A and Nvidia L40A nodes. NEMO1 is being phased out, with limited resources available until May 31st. Users are encouraged to transition to NEMO2 and consult the wiki for details.

NEMO2 uses Miniforge for conda environments, offering a streamlined setup with conda-forge as the default repository. The Miniforge module auto-initializes conda, simplifying environment activation without modifying shell profiles.

The AMD Genoa, Machine Learning and AI partitions for NEMO2 were delivered on December 4th. The acceptance of the storage has been delayed, so that NEMO2 could not yet start this year. However, calculations with the Milan nodes in NEMO1 are still possible.