Exercise: Using the environment modules to setup data-analysis software

In this exercise we will learn few useful commands for setting up data-analysis software in the cluster using the environment modules. Environment modules are helpful in organising software, and managing environment variables required by running the software.

The tasks below use the software R to illustrate the general idea that is applicable to setup other data-analysis software installed in the cluster.

Note

DO NOT just copy-n-paste the commands for the hands-on exercises!! Typing (and eventually making typos) is an essential part of the learning process.

Tasks

List the configured software

The following command is used to check what are software currently configure/setup in your shell environment:
```
$ module list
Currently Loaded Modulefiles:
1) cluster/1.0      3) matlab/R2018b    5) freesurfer/6.0
2) project/1.0      4) R/3.5.1          6) fsl/6.0.0
```
Configured software is listed in terms of the loaded modules.

You probably notice a message similar to the one above in the terminal after you login to the cluster’s access node. This message informs you about the pre-loaded environment modules. It implies that your bash shell has been configured with proper environment variables (e.g. PATH) for running those software/version right away after the login.
List available software
```
$ module avail
```
Environment modules for the software are organised in software names and versions.
List available versions of R
```
$ module avail R
```
You may replace R with matlab, freesurfer or fsl to see versions of different software.
Show the changes in environment variables w.r.t. the setup for R version 3.2.2
```
$ module show R/3.2.2
```
Check current value of the $R_HOME environment variable
```
$ echo $R_HOME
/opt/R/3.1.2
```
As the default R version, the $R_HOME variable is set to point to version 3.1.2.
Setup the environment for R version 3.2.2

Firstly, unload the default R with
```
$ module unload R
```
, and load the specific R version with
```
$ module load R
```
Following to it, check the $R_HOME variable again, it should be pointed to a directory where the version 3.2.2 is installed. You should be ready to use R version 3.2.2 in the cluster.
```
$ echo $R_HOME
```
Don’t like 3.2.2 and want to switch to 3.3.1 … Do you know how to do it?