executorlib.standalone.command.get_cache_execute_command

executorlib.standalone.command.get_cache_execute_command#

executorlib.standalone.command.get_cache_execute_command(file_name: str, cores: int = 1, backend: str | None = None, exclusive: bool = False, openmpi_oversubscribe: bool = False, pmi_mode: str | None = None) → list[source]#

Get command to call backend as a list of two strings

Parameters:

  • file_name (str) – The name of the file.

  • cores (int, optional) – Number of cores used to execute the task. Defaults to 1.

  • backend (str, optional) – name of the backend used to spawn tasks [“slurm”, “flux”].

  • exclusive (bool) – Whether to exclusively reserve the compute nodes, or allow sharing compute notes. Defaults to False.

  • openmpi_oversubscribe (bool, optional) – Whether to oversubscribe the cores. Defaults to False.

  • pmi_mode (str) – PMI interface to use (OpenMPI v5 requires pmix) default is None

Returns:

List of strings containing the python executable path and the backend script to execute

Return type:

list[str]