General Jobs

The General job submission page enables batch workload processing for arbitrary commands, containerized workloads, or template-based jobs.

Step 1: Job Scheduler Options

Configure the resources required for your job:

• Partition — select the SLURM partition (e.g., hpc).
• Job Name — assign a name to your job (e.g., My_Test_Job).
• CPUs per Task — specify the number of CPU cores (e.g., 2).
• Working Directory — auto-derived from your job root. Click the folder icon to open it in File Manager.

Expand Advanced Options for additional fields: Number of Nodes, Memory, Wall Time, etc.

Step 2: Execution Method

Choose an execution method using the tabs below the scheduler options:

Native

• Command — enter the shell command(s) to execute (required). Click the template button to load a saved command template.
• GUI App — enable to launch a graphical application via VNC instead of a batch job.

Container (Apptainer)

• Container Path — path to an Apptainer/Singularity .sif image (required). Use the Select button to browse available containers.
• Exec Command — command to run inside the container.
• Environment Variables — key-value pairs exported inside the container. Each variable has an enable/disable toggle.
• Mount Directories — host directories to bind-mount into the container.
• Use srun — enable for multi-node MPI applications.

Step 3: Environment Variables (Native mode)

Add key-value pairs to export as environment variables before the command runs. Click Add to add a row, and use the toggle to enable or disable individual variables.

Step 4: Submit

1. Optionally check Redirect to Job Monitor after submission.
2. Click Submit to send the job to the SLURM scheduler.
3. Monitor the job on the Job Monitor page or check output on the Job Logs page.