Versions Compared

Old Version 5

changes.mady.by.user Anna Battenhouse

Saved on

compared with

New Version Current

changes.mady.by.user Anna Battenhouse

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

Common Resources

Testing GPU access

Two Python scripts are located in /stor/scratch/GPU_info that can be used to ensure you have access to the server's GPUs. You can run them from the command line like this:

  • Tensor Flow
    • python3 /stor/scratch/GPU_info/tensorflow_example.py
  • PyTorch
    • python3 /stor/scratch/GPU_info/pytorch_example.py (not yet working)

If GPUs are available and accessible, the output generated will indicate they are being usedTo verify GPU access, use either the rocm-smi (AMD GPU servers) or nvidia-smi (NVIDIA GPU servers).

GPU-enabled software

See server-specific pages for how to run TensorFlow, PyTorch and AlphaFold and GROMACS.

Sharing resources

Since there's no batch system on BRCF POD compute servers, it is important for users to monitor their resource usage and that of other users in order to share resources appropriately.

...