Page History

240x

prod-[001-240]

smp, mpp

• 2x AMD Rome Epyc 7702 (64 cores each) → 128 cores

• 256GB RAM

• internal storage:

  • /tmp: 314GB NVMe

4x

fat-00[1-4]

fat

• 2x AMD Rome Epyc 7702

• 4TB RAM

• internal storage:

  • /tmp: 6.5TB NVMe

1x

gpu-001

gpu

• 2x AMD Rome Epyc 7702

• 1TB RAM

• internal storage:

  • /tmp: 3TB NVMe

  • /scratch: 6.3TB

• 2x AMD Rome Epyc 7702

• 1TB RAM

• internal storage:

  • /tmp: 3TB NVMe
  • /scratch: 6.3TB

• 2x Nvidia A40 GPU (48GB)

1x

gpu-002

gpu

• 4x Nvidia A100 GPU (80GB)

A comparison of the two different GPUs can be found here:
https://askgeek.io/en/gpus/vs/NVIDIA_A40-vs-NVIDIA_A100-SXM4-80-GB
The saying is: 

• How big are your models? Very, very big ⟹ A100

• Do you mainly work with mixed precision training (TensorFloat-32)? ⟹ A100

• Is FP32 more important? ⟹ A40

• Is FP64 more important? ⟹ A100

1x

gpu-002

gpu

• 2x AMD Rome Epyc 7702

• 1TB RAM

• internal storage:

  • /tmp: 3TB NVMe
  • /scratch: 6.3TB

• 4x Nvidia A100 GPU (80GB)