Versions Compared

Key

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

...

  • To Do
    • Change slurm so that nodes come up properly after a reboot instead of "unexpectedly rebooted"
    • SLURM runs two jobs that each ask for all the memory (#SBATCH --mem=0) on the same node.  That seems wrong.  May need to look info partition oversubscription settings.