OnDemand

From Research Computing Center Wiki
Revision as of 15:24, 23 August 2021 by Bcarpenter (talk | contribs) (Basic page complete)
Jump to navigation Jump to search

Open OnDemand (OOD)

Open OnDemand is a web-based service provided by GACRC which provides users an alternative and/or complimentary method to access the HPC Clusters.

  • It helps computational researchers and students efficiently utilize remote computing resources by making them easy to access from any device.
  • It helps computer center staff support a wide range of clients by simplifying the user interface and experience.

Throughout this documentation, the Open OnDemand application may be referenced by the acronym "OOD".

OnDemand Integration with Clusters
Cluster Phase URL
Sapelo2 Beta Testing https://ondemand.gacrc.uga.edu
Teach Future N/A

How to access OOD

Note: You must be connected to the UGA Network, either directory or by VPN, to access the OOD web page.

Browser Requirements/Notes

To have the best experience using OnDemand, use the latest versions of Google Chrome, Mozilla Firefox or Microsoft Edge. Use any modern browser that supports ECMAScript 2016.

Google Chrome has the widest range of support since the Shell App uses hterm.js which is supported officially by Google. Chrome currently is the only web browser that natively supports the copy and paste functionality in noVNC. Other browsers can do copy and pasting manually through the noVNC tool drawer.

Safari users may have issues when using the Shell App or noVNC.

There is no IE 11 support.

Main Page Layout

Example main page after login

After successful login the main page will be displayed.

A list of the different sections of the main page:

  • The Top Menu: See the individual documentation sections for each drop down: Files, Jobs, Clusters, Interactive Apps, My Interactive Sessions
  • Announcement Message: Shows important announcements only.
  • Message of the Day: Shows recent updates or notices about the OnDemand Service
  • Cluster Graphs: Shows Overall Node Utilization and Job Status on the cluster.