Scheduler Directives

Scheduler Specific Directives

For directives in your job script, it is recommended to use the Torque or Slurn specific directives and if there is a directive that is not available with CCQ, you may use the #CC directives listed below to accomplish your job needs.

Torque Specific Directives

The following options are Torque directives understood by CCQ. Only one directive must appear per line.

• Option / #PBS Directive	Description
  -e {stderr_file_location}	The path to the file where you want the Standard Error from your job to be written. The default location is the directory where ccqsub was invoked. The name of the file will be the job name combined with the job id on the machine the job was submitted from.
  -l {specification}	The specification in the format given by Torque’s documentation. CCQ understands only the mem, ncpus, and nodes directives which control instance type. The select directive is not a Torque directive, and an error will be produced if it is used.
  -o {stdout_file_location}	The path to the file where you want the Standard Output from your job to be written. The default location is the directory where ccqsub was invoked. The name of the file will be the job name combined with the job id on the machine the job was submitted from.

Slurm Specific Directives

The following options are Slurm directives understood by CCQ. Multiple directives on one line are permissable, however the entire line will be ignored if an unknown directive is encountered anywhere on it. Other lines will be processed normally.

• Option / #SBATCH Directive	Description
  -D {job_work_dir}	This specifies the path to the directory to use as the job’s working directory.
  -N {number_of_instances_requested}	This specifies the number of instances to start.
  -e {stderr_file_location}	The path to the file where you want the Standard Error from your job to be written. The default location is the directory where ccqsub was invoked. The name of the file will be the job name combined with the job id on the machine the job was submitted from.
  -o {stdout_file_location}	The path to the file where you want the Standard Output from your job to be written. The default location is the directory where ccqsub was invoked. The name of the file will be the job name combined with the job id on the machine the job was submitted from.
  –mem={memory}	This specifies the amount of memory per node. A number without suffix is interpreted to be the number of megabytes. The suffixes K, M, G, and T are understood.
  –ntasks-per-node={number_of_tasks_per_node}	This specifies the number of CPUs per node.