RStudio | Research Computing & Informatics

RStudio Server gives you the full RStudio interface in your browser, running on a cluster compute node. Write and run R scripts, make plots, and use the built-in terminal, with more memory and cores than a laptop and nothing to install.

On this page: Where it runs · Start a session · Launch settings · Installing packages · Files and sharing · Troubleshooting

Where it runs

Cluster	R versions	Runs on
SeaWulf	R 4.5.1 or 4.3.2, chosen on the form	Shared CPU queues (40-core or 96-core nodes)
NVwulf	Set by the app	B40 GPU nodes
ClinWulf	Set by the app	CPU or GPU queues, for work with PHI

To check the version inside a session, type R.version.string in the console.

Start a session

Sign in to your cluster's OnDemand portal with your NetID and Duo.
Open Interactive Apps and choose RStudio Server.
Fill in the form (see below) and click Launch.
When the card shows Running, click the connect button.
When you're done, save your scripts, close the tab, and click Delete on the session card.

Launch settings

Setting	SeaWulf	NVwulf	ClinWulf
Queue	`short-40core-shared` (up to 4 hours) for most work; long and extended shared queues for more	`b40x4` (up to 8 hours) or `b40x4-long` (up to 48 hours)	A CPU queue unless you need a GPU
Cores	1 to 4; more only if you use parallel packages	1 to 4	1 to 4
Memory	2 to 164 GB; about twice the size of your data is a good start	8 to 480 GB	Type it, for example `16G`
Extras	R version, and Additional modules for packages that need system libraries	GPUs (usually 1)	GPUs, typed as type and count (for example `a100:1`), only on GPU queues

The RStudio Server launch form with queue, cores, time, memory, and R version — The RStudio Server launch form (SeaWulf).

A running RStudio session card with the connect button — A running session, ready to connect.

Installing packages

Install packages the usual way. The first time, R asks to create a personal library in your home folder; answer yes.

install.packages("tidyverse")

Packages belong to one R version. If you switch versions on the form, reinstall the packages you need.
If an install fails with a missing library such as hdf5, libpng, or cmake, delete the session and relaunch with that module in Additional modules (SeaWulf), for example hdf5/1.12.1. Find module names with module avail in a terminal.
Bioconductor packages install with BiocManager::install() after install.packages("BiocManager").
For a reproducible project, renv keeps a separate library per project.

Files and sharing

RStudio opens in your home folder. To reach scratch or project space from the Files pane, add a shortcut once from the RStudio Terminal tab, for example on SeaWulf:

ln -s /gpfs/scratch/$USER ~/scratch
ln -s /gpfs/projects/<group> ~/project

RStudio sessions are private; there's no link to share a running session. Work together through Git (RStudio's Git pane works as usual) or a shared project folder.

RStudio in the browser with the editor, console, environment, and files panes — RStudio Server running in the browser.

Troubleshooting

R crashes or says "cannot allocate vector"

You ran out of memory. Save your work, delete the session, and relaunch with more memory.

RStudio opens with an old, broken workspace

RStudio restores your last session. Turn this off under Tools, Global Options, General: untick restoring .RData and set saving the workspace to Never. If it still won't open, rename ~/.local/share/rstudio from a terminal and relaunch.

A package won't install

Read the error for a missing system library, then relaunch with that module added (see above). If you're stuck, open a ticket with the package name and the error.

My session stays Queued

Try a shorter queue, fewer cores, or less memory.

Applies to SeaWulfNVwulfClinWulf