HTML R Shell
Switch branches/tags
Nothing to show
Clone or download
Latest commit df059e8 Jun 8, 2018


CRAN Build Status

googleComputeEngineR provides an R interface to the Google Cloud Compute Engine API, for launching virtual machines. It looks to make the deployment of cloud resources for R as painless as possible, and includes some special templates to launch R-specific resources such as RStudio, Shiny, and OpenCPU with a few lines from your local R session.

See all documentation on the googleComputeEngineR website

TL;DR - Creating an RStudio server VM

  1. Configure a Google Cloud Project with billing.
  2. Download a service account key JSON file.
  3. Put your default project, zone and JSON file location in your .Renviron.
  4. Run library(googleComputeEngineR) and auto-authenticate.
  5. Run vm <- gce_vm(template = "rstudio", name = "rstudio-server", username = "mark", password = "mark1234") (or other credentials) to start up an RStudio Server.
  6. Wait for it to install, login via the returned URL.

A video guide to setup and launching an RStudio server has been kindly created by Donal Phipps and is available at this link.

<iframe width="560" height="315" src="" frameborder="0" allowfullscreen></iframe>

Thanks to

  • Scott Chamberlin for the analogsea package for launching Digital Ocean VMs, which inspired the SSH connector functions for this one.
  • Winston Chang for the harbor package where the docker functions come from. If harbor will be published to CRAN, it will become a dependency for this one.
  • Henrik Bengtsson for help in integrating the fantastic future package that allows asynchronous R functions run in GCE clusters.
  • Carl Boettiger and Dirk Eddelbuettel for rocker that Docker containers some of the R templates used in this package.


CRAN version:


Development version:

if (!require("ghit")) {

cloudyr project logo