helixforge parallel example-sbatch

Write a copy-paste SBATCH wrapper that runs a tasks file on one node.

Convenience starting point for the scheduler-agnostic path: it runs the whole tasks.txt (from parallel tasks) inside a single allocation, fanning out across $SLURM_CPUS_PER_TASK cores via HyperShell or GNU Parallel. Edit the partition / time / resources for your cluster. For a true one-task-per- chunk Slurm array driven off a plan, use reconcile --hpc slurm instead.

Options

Option	Description
-o, --output	Where to write the example SBATCH script. (required)
--executor	Run the task file with HyperShell or GNU Parallel. (choices: hypershell/parallel; default: hypershell)

Examples

  # HyperShell wrapper (default)
  helixforge parallel example-sbatch -o run.sbatch

  # GNU Parallel wrapper
  helixforge parallel example-sbatch -o run.sbatch --executor parallel