Super Computers
At Nazarbayev University, we recognize the essential role HPC plays in modern research, innovation, and problem-solving across disciplines. Our facilities combine advanced technologies with robust infrastructure to support a wide range of demanding computational tasks.
NU Super Computing Cluster
- Three clusters: Irgetas, Shabyt, Muon
- Irgetas: 10 CPU-only nodes with two 96-core AMD EPYC 9684X CPUs each, and 6 GPU nodes with two 96-core AMD EPYC 9454 CPUs and four NVIDIA H100 GPUs each
- Shabyt: 20 CPU-only nodes with two 32-core AMD EPYC 7502 CPUs each, and 4 GPU nodes with two 32-core AMD EPYC 7452 CPUs and two NVIDIA V100 GPUs each
- Muon: 10 compute nodes each with a single 14-core Intel Xeon CPU
- Up to 768 GB RAM per node (Irgetas H100 partition)
- Home directory quotas: 400 GB (Irgetas), 100 GB (Shabyt), 250 GB (Muon)
- Additional group storage: 1 TB allocation per research group on Irgetas, Shabyt, and Muon
- Job scheduling via SLURM with dedicated CPU and GPU partitions
- Max job durations range from 2 days (Shabyt NVIDIA partition) up to 14 days (Shabyt CPU and Muon HPE partitions)
