Table of Contents |
---|
For complete SLURM documentation, see https://slurm.schedmd.com/. Here we only show simple examples with system-specific instructions.
HAL Job Queues
Partition Name | Priority | Max Walltime | Max Nodes/Job | Description |
---|---|---|---|---|
debug | high | 12 hrs | 1 | designed to access 1 GPU to run debug or short-term job |
solo | normal | 72 hrs | 1 | designed to access 1 GPU to run long-term job |
batch | normal | 72 hrs | 16 | designed to access up to 64 GPUs to run parallel job |
...