How to work with the IFB cluster

You're going to use Open OnDemand, a browser-based portal to all resources within the IFB cluster.

Sign in on IFB cluster Open OnDemand

In a web browser, connect to https://ondemand.cluster.france-bioinformatique.fr// and log in using your IFB username and password.

Open OnDemand

Clic on the "Grant Access" button.

Open OnDemand Access

Internal Server Error

If you get the error message "Internal Server Error" after grant access, this is a problem with the browser's cookies. Switching to a private browser window solves the problem.

Bookmark this URL in your browser for easy retrieval.

Using a JupyterLab session to access Unix terminal

Clic on the "JupyterLab" tile

Open OnDemand Applications

Select your server options

On the following screen you can select the option used by our JupyterHub server during the practical. Two nodes have been booked for our training session on IFB server. To build your server on these nodes select the following options:

  • Reservation: 2528_fgat
  • Account: 2528_ens_master2lf_fgat
  • Partition: fast
  • CPU: 2
  • Memory: 2G
  • GPUs: No GPU
  • Number of hours: 3

JupyterLab server options

Clic on "Launch" and wait for the JupyterLab session to be created. Once running, clic on "Connect to Jupyter" to access the server.

JupyterLab connect

Start a terminal session

Once your server has been launched, you will be taken to the application launcher page.

JupyterLab launcher

You can click on the "Terminal" tile in order to open a terminal window.

JupyterLab Terminal

Using an Rstudio environment

Clic on the "RStudio Server" tile

Open OnDemand Applications

Select your server options

On the following screen you can select the option used by our RStudio server during the practical. Two nodes have been booked for our training session on IFB server. To build your server on these nodes select the following options:

  • Reservation: 2528_fgat
  • Account: 2528_ens_master2lf_fgat
  • Partition: fast
  • CPU: 2
  • Memory: 3G
  • GPUs: No GPU
  • Number of hours: 3

RStudio server options

Clic on "Launch" and wait for the RStudio session to be created. Once running, clic on "Connect to RStudio Server" to access the server.

RStudio connect

Rstudio interface

You can get more information using the IFB-core Cluster documentation and the dedicated Open On Demand pages.