02-rstudio.Rmd

# RStudio Workbench

RStudio Workbench (RSW) is running on the main node of the cluster.
It aims to be the central go-to place for scripts and analysis development.

RSW is running in a Docker container which is running on Ubuntu 20.04.
This means that if you open a terminal in RSW, you are in an Ubuntu environment.

```{block, type='rmdcaution'}
This is different to the environment if you `ssh` directly onto the server, which is running on a CentOS7 system.
```

All packages installed in RSW are linked against Ubuntu 20.04 system libraries which are already installed in the RSW container.

RSW is running on the main node which has 32 cores and 124 GB RAM.
This machine is shared with multiple people, so please ensure that you do not use too much memory and clean up from time to time.
RSW shows you your memory usage in the "Environment" pane of the IDE.
Besides, you can go to the terminal and execute `htop` to check the current load on the server.

## SLURM

RSW is connected to the SLURM scheduler and is able to add jobs to the SLURM queue.

## Python

RSW ships with support for JupyterLab and JupyterNotebooks.
You an launch a Jupyter session by selecting it in the drop-down of a new session instead of "RStudio".

## VSCode

RSW >= 1.4 ships with a full VSCode integration.
You an launch a VSCode session by selecting it in the drop-down of a new session instead of "RStudio".

# Shiny Server

Shiny server is running on port 3838 with user support.
This means users can place apps in `$HOME/Shinyapps/` and they will be deployed as `http://edi.geogr.uni-jena.de:3838/<username>/<appname>`.

Note the `http` (`https` is only available in the Pro version).

Exemplary apps:

- http://edi.geogr.uni-jena.de:3838/patrick/hyperspectral/
- http://edi.geogr.uni-jena.de:3838/jannes/cluster_map/