Using R in batch

Batch mode¶

There are two senses of 'batch' at play: One is R's sense of batch, and the other is the sense of 'batch job'. On the cluster, you may need to run R in a batch job, so you need some way to tell R to run the commands in a file.

One way is to use R's batch mode, with which R's output will go to a file rather than the screen. The other way is to use Rscript, which prints output to standard out, and in the context of a batch job, that will be the job's output file.

Using R's batch mode¶

For longer jobs, or if you need to run many similar R programs, you will want to run those in a batch job. There are two ways to run an R script as if it were a program, one uses the R command and the the other uses Rscript.

For the following examples, you should make a test directory and copy our example R script to it using these commands (the $ is the prompt; do not type it).

$ mkdir Rtest
$ cd Rtest
$ cp /gpfs1/sw/examples/R/iris.R .
$ module load R

Using `R CMD BATCH`¶

The first method of running an R script in batch is to use the usual R but to add the CMD parameter. In its simplest form that looks like this:

$ R CMD BATCH iris.R

That will instruct R to run the commands in iris.R and put the output into a file called iris.Rout in the same directory. If you want to rename the output file, add the desired output file name to the command, as in

$R CMD BATCH iris.R iris.txt

Using `Rscript`¶

Rscript sends output to standard output, so, when you run it from an interactive prompt, the output appears on screen.

$ Rscript iris.R
  Sepal.Length    Sepal.Width     Petal.Length    Petal.Width   
 Min.   :4.300   Min.   :2.000   Min.   :1.000   Min.   :0.100  
 1st Qu.:5.100   1st Qu.:2.800   1st Qu.:1.600   1st Qu.:0.300  
 Median :5.800   Median :3.000   Median :4.350   Median :1.300  
 Mean   :5.843   Mean   :3.057   Mean   :3.758   Mean   :1.199  
 3rd Qu.:6.400   3rd Qu.:3.300   3rd Qu.:5.100   3rd Qu.:1.800  
 Max.   :7.900   Max.   :4.400   Max.   :6.900   Max.   :2.500  
       Species  
 setosa    :50  
 versicolor:50  
 virginica :50

If you use Rscript in a batch job, output will go to the Slurm output file instead of the screen.

Using R in a batch job¶

To use R from a (Slurm) job, you need to create a job script. Here is an example for you to put into a file called r-cmd.sbat in the Rtest folder.

#!/bin/bash

####  Slurm preamble to define the job
#SBATCH --job-name=R_example
#SBATCH --nodes=1
#SBATCH --ntasks-per-node=1
#SBATCH --cpus-per-task=1
#SBATCH --mem=4gb
#SBATCH --time=5:00
####  End preamble

#### Commands to run in the job

##  Clear all modules and load only those needed for the job
module purge
module load R

##  Run your commands
echo "I ran from $(pwd) on $(date)"
R CMD BATCH iris.R

The echo line prints some information to the standard output; otherwise the Slurm output file would be empty unless error messages were printed. R CMD BATCH doesn't normally produce any output. To submit the job, use

$ sbatch r-cmd.sbat
Submitted batch job 12345

The output from this job will be in a file called slurm-12345.out in the directory from which you submitted the job, and it will contain just the results of running the echo command.

$ cat slurm-12345.out
I ran from /gpfs1/home/u/s/user/Rtest on Tue May 20 12:05:53 PM EDT 2025

and the R output will be in iris.Rout.

Copy r-cmd.sbat to rscript.sbat and change the last line to read

Rscript iris.R

and then submit it with

$ sbatch rscript.sbat
Submitted batch job 12346

When it completes, both the output from the echo command and from Rscript will be in slurm-12346.out.

$ cat slurm-12346.out
I ran from /gpfs1/home/u/s/user/Rtest on Tue May 20 12:15:55 PM EDT 2025
  Sepal.Length    Sepal.Width     Petal.Length    Petal.Width   
 Min.   :4.300   Min.   :2.000   Min.   :1.000   Min.   :0.100  
 1st Qu.:5.100   1st Qu.:2.800   1st Qu.:1.600   1st Qu.:0.300  
 Median :5.800   Median :3.000   Median :4.350   Median :1.300  
 Mean   :5.843   Mean   :3.057   Mean   :3.758   Mean   :1.199  
 3rd Qu.:6.400   3rd Qu.:3.300   3rd Qu.:5.100   3rd Qu.:1.800  
 Max.   :7.900   Max.   :4.400   Max.   :6.900   Max.   :2.500  
       Species  
 setosa    :50  
 versicolor:50  
 virginica :50