Installing and Managing R Packages on Quest
Installing and managing R packages on Quest.
R packages are installed, whether system-wide or locally in your own
directory, for specific versions of R. If you switch versions of R, you
may need to reinstall packages you use. However, packages only need to
be installed once on Quest for each version of R. Once installed, a
package is available across all log in, compute, and analytics nodes on Quest for that version of R.
A limited number of R packages have been installed centrally for each version of R. These packages can be used without downloading and installing them first. You can get a list of the currently installed packages with the command:
Users can install additional R packages from CRAN by launching RStudio and using the Packages tab in the bottom-right window, or by running the R command line console and using the install.packages command:
with a list of the packages you need. Packages should generally be installed under your home directory. They will be available for you to use across Quest nodes.
The first time you install an R package on Quest, you may see an error like:
Warning in install.packages("glmnet", repos = "http://cran.r-project.org") : 'lib = "/hpc/software/R/3.3.1/lib64/R/library"' is not writable Would you like to use a personal library instead? (y/n)
Answer "y" to use a personal library. Then it will ask you:
Would you like to create a personal library ~/R/x86_64-pc-linux-gnu-library/3.3 to install packages into? (y/n)
Answer "y" again. The installation should then proceed successfully. This will install the packages in your home directory.
Setting the Repository
If you get an error message indicating that a particular package isn't available for a certain version of R, or you get another error related to the package repository, you may need to explicitly set the mirror from which you want R to download the package. List of CRAN mirrors. Choose an http (instead of https) mirror. You can set the mirror before trying to install packages with the command:
or as part of the install.packages call with the repos argument:
install.packages(c("packagename1", "packagename2"), repos="http://cran.wustl.edu")
For Use on Login and Compute Nodes
Some packages compile underlying code as a part of the installation. To install such packages, you may need to load a newer version of the gcc compiler before starting R and installing the package. Loading a different compiler should NOT be done for packages you intend to use on the Analytics nodes, as the updated gcc modules are not available on those nodes.
For example, if while installing a package, you see an error like:
configure: WARNING: Only g++ version 4.6 or greater can be used with...
try loading the module for a newer version of gcc; For example, after logging into Quest:
module load gcc/5.1.0
module load R/3.3.3
then in the R console, issue the package installation commands again.
For Use on Analytics Nodes
R packages intended for use on the Analytics nodes should be installed from within an R session running in RStudio in the web browser on the Analytics nodes. If you encounter compilation or installation errors for a package, please contact firstname.lastname@example.org with the details for further assistance.
R packages installed via the Quest login or compute nodes with the module R/3.5.1 will also be available on the Analytics nodes.
Some compiled packages cannot be successfully installed by individual users. Please contact email@example.com if you run into trouble with a particular package.
See the R help page for install.packages for more options, such as installing local (user-created or downloaded) packages or changing the directory where packages are installed. The latter can be used to install packages in your project space instead of your home directory, which might be useful if others on the allocation also need to access the packages. To install packages from GitHub, use the install_github function in the devtools package or use the githubinstall package.
CRAN Task Views
are collections of packages used frequently in particular fields. They
allow you to install all of the packages associated with the task view
at once. See the R documentation for more details and the commands to
Please contact firstname.lastname@example.org for additional help with R package installation.