Open OnDemand is a web portal that provides rich access to HPC systems. It allows users to view, edit, upload and download files; create, edit, submit and monitor jobs; run interactive GUI applications such as Jupyter Notebook and Rstudio; and use the Linux shell--all via a web browser and with minimal knowledge of Linux and scheduler commands.
This brief document provides an overview of the features available from the Open OnDemand interface to the ODU’s Wahab cluster.
To connect, point your web browser to
https://ondemand.wahab.hpc.odu.edu/
and log in with your ODU MIDAS username/password and your multi-factor authentication device. You will be presented with a dashboard page that has a top menu bar like this:
This menu provides access to the power and capabilities within the Open OnDemand interface. Each capability is provided through a single "web app", which runs on a separate tab on your browser and has a fixed URL. For example, the Dashboard’s URL is always https://ondemand.wahab.hpc.odu.edu/pun/sys/dashboard .
The Files menu allows one to view and operate on files available on the cluster (by default, starting in the user's home directory). The File Management and Transfer help page provide details on its use. The features available on OnDemand’s File Explorer are similar to those available in a GUI file manager on the remote desktop interface. A user can also edit text files, which is a very welcome feature for those not familiar with command-line editors like nano and vi. File Explorer can also be used to upload files to, or download files from, the cluster using, much like WinSCP or CyberDuck.
A user can use the Jobs menu to create, edit, submit, and monitor jobs on the cluster.The Job Management help page provides more information on the use and features. The Active Jobs submenu item serves as a GUI alternative to SLURM scheduler commands and allows us to write SLURM batch scripts with the help of pre-defined templates (Job Composer menu item).
The Clusters menu provides access to the Linux shell on Wahab's login node. For security reasons, you will be asked your MIDAS password again before accessing the shell.
Hint: If necessary, you could use
ssh turing.hpc.odu.edu
from the shell to access Turing login node.
The Interactive Apps menu allows you to launch interactive jobs on Wahab cluster. Presently, it has two apps: JupyterLab, Rstudio and a Virtual Desktop. Please refer to the relevant documentations below: