Job Management

The Jobs menu in Open OnDemand gives you two tools for working with SLURM batch jobs without opening a terminal: a live view of your active jobs, and a form-based job composer for writing and submitting scripts.

Active Jobs

Jobs → Active Jobs shows a live table of all SLURM jobs currently associated with your account — running, pending, completing, and recently finished.

Reading the jobs table

Each row represents one job. The columns are:

Column	Description
ID	The SLURM job ID assigned at submission
Name	The job name from #SBATCH --job-name
User	Your username
Account	The SLURM account charged for this job
Partition	The partition (queue) the job is running or waiting in
Status	Current status: RUNNING, PENDING, COMPLETED, FAILED, etc.
Time Used	How long the job has been running
Time Limit	The wall-time limit requested
Nodes	Number of nodes allocated
Reason	For pending jobs, why SLURM is holding the job (e.g., Resources, Priority)

ACE HPC partitions

Jobs on ACE HPC are submitted to one of the following SLURM partitions:

Partition	Intended use	Default time limit	Max time limit
shared	Shared CPU compute	1 hour	Unlimited
extended	Exclusive CPU usage	1 hour	Unlimited
gpu	Jobs requiring a GPU	1 hour	48 hours

Verify partition details

Partition names, time limits, and resource caps are subject to change. Run sinfo from the Shell Access or check with the systems team for the current configuration.

Filtering and sorting

Use the filter box at the top of the table to search for a specific job by name or ID. Click any column header to sort the table by that column.

Cancelling a job

To cancel a running or pending job:

1. Find the job in the table
2. Click the red delete icon (trash can) in the rightmost column of that row
3. Confirm the cancellation in the prompt that appears

This is equivalent to running scancel <job_id> in the terminal.

---

Job Composer

Jobs → Job Composer lets you create, edit, and submit SLURM batch scripts entirely through the browser — no terminal needed.

Creating a new job

1. Click New Job → From Default Template (or choose a saved template if one exists)
2. A new job entry is created and opens in edit mode

The job composer shows three panels:

  ┌──────────────────────┬───────────────────────────────────────┐
  │  Job list (left)     │  Job Script Editor (right)            │
  │                      │                                       │
  │  ▶ My job 1          │  #!/bin/bash                          │
  │  ▶ My job 2          │  #SBATCH --job-name=my_job            │
  │  [+ New Job]         │  #SBATCH --partition=normal           │
  │                      │  #SBATCH --nodes=1                    │
  │                      │  #SBATCH --ntasks=1                   │
  │                      │  #SBATCH --mem=4G                     │
  │                      │  #SBATCH --time=01:00:00              │
  │                      │                                       │
  │                      │  module load python/3.10              │
  │                      │  python my_script.py                  │
  └──────────────────────┴───────────────────────────────────────┘

Editing the job script

The right panel is a text editor where you write or paste your SLURM script. A minimal script for ACE HPC looks like this:

#!/bin/bash
#SBATCH --job-name=my_analysis
#SBATCH --partition=normal
#SBATCH --nodes=1
#SBATCH --ntasks=4
#SBATCH --mem=16G
#SBATCH --time=02:00:00
#SBATCH --output=job_%j.out
#SBATCH --error=job_%j.err

module load python/3.10
python my_analysis.py --threads 4

See Writing Job Scripts for a full explanation of every directive.

Setting the job working directory

Below the script editor, there is a Job Directory field. This controls where SLURM will create the output (.out) and error (.err) files. Set this to the directory where your input data and scripts live, for example /etc/ace-data/<your-username>/projects/my_project.

You can also click Select Path to browse your filesystem and choose the directory graphically.

Submitting the job

Once your script is ready:

1. Click Save to save the script
2. Click Submit — the green button in the toolbar

The job will be queued in SLURM. You will see a confirmation message with the assigned job ID, and the job will appear immediately in Active Jobs.

Copy from terminal workflows

If you already have a working SLURM script on the cluster, the easiest approach is to paste it directly into the Job Composer editor rather than retyping it. Use the Shell Access to cat your existing script, then copy and paste.

Managing saved jobs

The left panel lists all jobs you have composed. You can:

• Click a job to open and edit it
• Copy a job to use it as a template for a new one
• Delete a job to remove it from the list (this does not cancel a running SLURM job — it only removes the script from the composer)