Page History
...
A short summary of the hardware available in the nodes:
Partition | NNodes | GPUs per node | GPU | GPU memory | Max time |
---|---|---|---|---|---|
normal | 1300 | 4 | GH200 | 96GB | 12 hours |
debug | 0 | 4 | GH200 | 96GB | 30 minutes |
Each node consists of 4xGH200 superchips. Each superchip is a unified memory system consisting of a Grace CPU and a Hopper GPU with a 900GB/s NVLINKC2C connect. The Grace CPUs share 512GB LPDDR5X memory. Each individual Hopper GPU has 96GB HBM3 memory with 3000GB/s read/write, totaling 896GB of unified memory available within each node.
More information on the available partitions can be found with with scontrol show partitions.
...