R Software environment
R is a widely used software environment with a focus on statistical computing and graphics.
Finding available R versions
Information on available R versions installed can be obtained with the module spider command on the LINUX command line:
which will get you an output similar to:
----------------------------------------------------------------------------
R:
----------------------------------------------------------------------------
Description:
R is a free software environment for statistical computing and
graphics.
Versions:
R/3.2.1-bare
R/3.2.1
R/3.2.3-libX11-1.6.3
R/3.2.3
R/3.3.1
R/3.3.2-libX11-1.6.3
R/3.4.0-X11-20170314
R/3.4.3-X11-20171023
R/3.5.0-X11-20180131
R/3.5.1-Python-2.7.15
R/3.5.1
R/3.6.0
R/3.6.2
R/3.6.3
R/4.0.0
R/4.1.0
R/4.1.2
Select the version that best fits your need and run module spider again to ask for detail on this specific version. So if we want to run R/4.1.2 we run (cut-n-paste is your friend):
This will give faily extensive output:
----------------------------------------------------------------------------
R: R/4.1.2
----------------------------------------------------------------------------
Description:
R is a free software environment for statistical computing and
graphics.
You will need to load all module(s) on any one of the lines below before the "R/4.1.2" module is available to load.
GCC/11.2.0 OpenMPI/4.1.1
Help:
Description
===========
R is a free software environment for statistical computing
and graphics.
More information
================
- Homepage: https://www.r-project.org/
Included extensions
===================
abc-2.1, abc.data-1.0, abe-3.0.1, abind-1.4-5, acepack-1.4.1, adabag-4.2,
ade4-1.7-18, ADGofTest-0.3, aggregation-1.0.1, AICcmodavg-2.3-1,
akima-0.6-2.2, alabama-2015.3-1, AlgDesign-1.2.0, AnalyzeFMRI-1.1-24,
animation-2.7, aod-1.3.1, ape-5.5, argparse-2.1.2, arm-1.12-2, askpass-1.1,
asnipe-1.1.16, assertive-0.3-6, assertive.base-0.0-9, assertive.code-0.0-3,
assertive.data-0.0-3, assertive.data.uk-0.0-2, assertive.data.us-0.0-2,
assertive.datetimes-0.0-3, assertive.files-0.0-2, assertive.matrices-0.0-2,
assertive.models-0.0-2, assertive.numbers-0.0-2, assertive.properties-0.0-4,
assertive.reflection-0.0-5, assertive.sets-0.0-3, assertive.strings-0.0-3,
assertive.types-0.0-3, assertthat-0.2.1, AUC-0.3.0, audio-0.1-8, aws-2.5-1,
awsMethods-1.1-1, b-a, backports-1.3.0, bacr-1.0.1, bartMachine-1.2.6,
...
The key information here is a list of prerequisite modules and a list of included R extensions. In this example you need to load the modules GCC/11.2.0and OpenMPI/4.1.1 before being able to load R.
Making R available and starting R
Once you have selected "your" version of R it is now time to load the R-module. Staying with our example of R 4.1.2, we have found out that it requires GCC/11.2.0 and OpenMPI/4.1.1. We load these and then load the R-module from the command line:
Now we can start R using the R command on the commandline or inside a batch script
Note
Resource intensive R jobs (compute time and/or memory consumption) have to use the compute nodes via the batch scheduler. Such jobs must not utilise login or frontend nodes.
Installing and using R extensions in your home space
Inside R you can use the R-function library() to display the available extensions. If your required extension is not available, you can install it in your own disk space. For that we recommend creating a directory in your home space, which you may name MyRextensions. This has to be done once for your account and one can store many extensions in that directory. The creation of this directoy can be accomplished at the LINUX command line utilising:
Once the directory has been created, we recommend the use of the R-function .libPaths() inside your R session. This will make R-functions such as library() and install.packages() use this directory:
MyRextensions is the first argument. After this, any call to install.packages() will install the extensions into the directory MyRextensions inside your diskspace. For instance, if we want to install the extension geosphere we have to run:
Example R session to install an extension
A session to install the extension geosphere on the existing directory ~/MyRextensions on the commandline may look as follows:
-bash-4.2$ R
R version 4.2.1 (2022-06-23) -- "Funny-Looking Kid"
Copyright (C) 2022 The R Foundation for Statistical Computing
Platform: x86_64-pc-linux-gnu (64-bit)
R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.
Natural language support but running in an English locale
R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.
Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.
> .libPaths(c('~/MyRextensions', .libPaths()))
> .libPaths()
[1] "/home/jhein/MyRextensions"
[2] "/sw/easybuild/software/R/4.2.1-foss-2022a/lib64/R/library"
> install.packages('geosphere')
...
> quit()
Save workspace image? [y/n/c]: n
The install.packages will create quite some output and ask you to select a CRAN mirror, which we ommitted in this guide. Note that the .libPaths() command is not necessary to the build; it just prints your library path so you can make sure that your personal extension directory has been added.
Using a user installed package
To use the package you have to make it available using the library() function of R. If it installed in your own space you have to tell R where to look for it.
Modifying your R-scripts
You can do this in your R-script by using .libPaths():
Using the environment variable R_LIBS
An alternative to modifying your R-scripts is to set the environment variable R_LIBS inside your SLURM submission script, prior to starting R:
Author: Joachim Hein
Last Updated: 2023-01-20