Using R on WTS

Using R on WTS

Connect to WTS and open the start menu. From there, you can choose between

  • Rstudio
  • CRAN R
  • Microsoft R Open

Which one should I use?

If you want an IDE, RStudio is the best choice as it includes an editor as well as a debugger and visualization tools. If you don't need an IDE, you can use either CRAN R or Microsoft R Open. CRAN R is the version most users are familiar with whereas Microsoft R Open includes some enhancements, most notably multi-threading capability.

RStudio

To use RStudio, just select it from the start menu on WTS. If you're new to RStudio, the Getting started with R and RStudio page at Coding Club has a good (but short) intro to using RStudio. The support page for Using the RStudio IDE also has documentation that covers a fairly wide range of topics that should be helpful in getting you started.

CRAN R

CRAN R is GNU software and is the version that most users of R are familiar with. With CRAN, you always have access to the latest packages as well as community support and knowledge.

Microsoft R Open

Microsoft R Open (MRO) is an enhanced R distribution formerly known as Revolution R Open that offers additional functionalities for the R language.

  • Total compatibility with (the same version of) CRAN R.
  • Multi-threaded capabilities
    • MRO includes multi-threaded math libraries, making it possible for many common R operations, such as matrix multiply/inverse and matrix decomposition, to compute in parallel, thereby reducing computation times.
  • Version Control
    • The fixed CRAN snapshot ensures that every user of a specific version of MRO will have access to the same packages and versions.
    • The checkpoint package allows you to take custom snapshots of your R configuration at any point in time. This makes it easy to share R code and replicate results using specific R package versions.
  • Uses MRAN to install snapshots of packages from CRAN, which allows for better reproducibility.
  • Compatible with R user interfaces, such as RStudio.

Although MRO offers a lot of benefits, there are also some drawbacks.

  • You may not be able to upgrade or install a package that depends on a newer release of R than that provided by MRO.
  • For the reproducibility to work, you (and everyone on your team) will need to fully buy in and use the checkpoint feature to lock package versions.