Skip to content

Open OnDemand

Open OnDemand provides browser-based access to CRC HPC resources, allowing you to run jobs, manage files, and launch interactive applications without using SSH.

Log In

KU Network

You must be on KU/KUMC network or KU anwyhere to access the OnDemand site

  1. Navigate to https://ondemand.crc.ku.edu
  2. Sign in with your KU credentials
  3. You'll be taken to the Open OnDemand dashboard.

Access Your Files

Select Files from the top menu to:

  • Browse your directories
  • Upload or download files
  • Create folders
  • Edit text files

Only for small transfers

For large data transfers, use SCP, SFTP, rsync, or Globus when available.

Terminal

Select Clusters from the top menu and then KU Community Cluster Shell Access.

This opens a terminal session on a CRC login node where you can:

  • Run Linux commands
  • Manage files
  • Submit Slurm jobs
  • Load software modules

Example:

module avail
squeue -u $USER

Not for processing

This login node is a simple VM, not for processing of files. SSH to hpc.crc.ku.edu if you want regular login node.

Launch Interactive Applications

Select Interactive Apps and choose an application such as:

  • Jupyter Notebook / JupyterLab
  • RStudio
  • VS Code
  • Interactive Desktop

Starting a Session

  1. Select an application.
  2. Choose the requested resources:
  3. CPUs
  4. Memory
  5. Time limit
  6. Click Launch.
  7. When the job starts, click Connect.

Reconnect to Running Sessions

If you close your browser or lose connection:

  1. Log back into Open OnDemand.
  2. Find your job under My Interactive Sessions.
  3. Click Connect.

Your application will continue running until the job ends or is deleted.

End a Session

When finished:

  1. Go to My Interactive Sessions
  2. Click Delete

This stops the job and releases allocated resources.

Job Composer

Select Jobs -> Job Composer to create and submit Slurm batch jobs through a web interface.

Example Slurm script:

#!/bin/bash
#SBATCH --job-name=test
#SBATCH --time=01:00:00
#SBATCH --cpus-per-task=4
#SBATCH --mem=8G
#SBATCH --partition=sixhour

module load python

python myscript.py

Monitor Jobs

Select Jobs -> Active Jobs to view:

  • Running jobs
  • Pending jobs
  • Job status
  • Runtime information

This provides a graphical alternative to:

squeue -u $USER