Graphical interfaces

Clusters are about batch processing: they provide tons of resources and there are lots of users, and all those scheduled as efficiently as possible. But sometimes, being able to see the computations as they come out with your own eyes helps (mainly in either development or debugging). You can, in fact, do this.

Initial notes

GUI

  • Many start with GUI

  • Ways this is possible on HPC too

  • RStudio/Jupyter/VSCode

  • GUI software via OOD or -X

X Forwarding

X Forwarding

  • ssh -X and ssh -Y

  • xeyes

  • Navigate to the R example directory

  • gnuplot -c gnuplot.plt

  • Sends the image of the GUI to your local machine and the input back to the remote machine.

  • Needs a fast network, otherwise get’s very laggy

Open On Demand

  • A more efficient way of running graphical applications on the cluster

  • The interface depends on the cluster

  • Not available on all clusters

Open On Demand

  • Open OOD

  • Look at the file system, verify it is the cluster

  • Open the Rstudio app

  • Open and edit the R script

Graphical User Interfaces

  1. Graphical User Interfaces

    • Find out what GUIs are available on the cluster

    • Is your favorite text editor available?

    • What other graphical applications are available? What would you use if they were?

    • Try uploading or downloading afile

X Forwarding

  1. X Forwarding

    • What is the difference between ssh -X and ssh -Y? (Do a web search and see what you can find.)

    • Try running a graphical application on the cluster and displaying it on your local machine