Software and package installation¶

Install software yourself¶

To build from source use a compiler module
We have several compiler versions from GNU and INTEL
check with: $ ml avail gcc and $ ml avail intel
Guide for compiling serial programs
Guide for compiling parallel programs
- Available combinations of compilers and parallel libraries

This guide might not work on all programs. Read the installation instructions for your program!

Download the program, with wget or by other means like git clone <https-URL to GITHUB repo>.
If the not cloning, unpack it with tar, gunzip or similar.

tar xvfz program.tgz

The below is more general again:

Read the installation instructions!
If Fortran or C or C++, load a compiler. Often you'll have less problems with gcc but intel gives faster code.

module load gcc

mkdir $HOME/glob/program_name
./configure --prefix=$HOME/glob/program_name
make
make test
make install

Try to find a test on the home page of the program or in the installation instructions and try to run it.

Python, R and Julia all have some centrally installed packages that are available from the modules.
R has a special module called R_packages, and some Machine Learning python packages are included in the python_ML_packages module.
If not found there you can try to install those by yourself.

Tip Python packages

Try Conda first directly on Bianca and PyPI on Rackham.
We have mirrored all major Conda repositories directly on UPPMAX, on both Rackham and Bianca. These are updated every third day.
If you want to keep number of files down, use PyPI (pip).
Also it is easier to get conflicting environments if using both Python module and Conda in parallel.

We have mirrored all major Conda repositories directly on UPPMAX, on both Rackham and Bianca. These are updated every third day.

Available Conda channels

On UPPMAX the module R_packages is an omnibus package library containing almost all packages in the CRAN and BioConductor repositories.
As of 2023-05-31, there were a total of 23100 R packages installed in R_packages/4.2.1.
- A total of 23109 packages were available in CRAN and BioConductor, and 23000 of these were installed in R_packages/4.2.1
- The additional 100 R packages available in this module were installed from the CRAN/BioConductor archives, or were hosted on github, gitlab or elsewhere.
Installing R packages

At UPPMAX there is a central library with installed packages.
This is good, especially when working on Bianca, since you then do not need to install via the wharf.
A selection of the Julia packages and libraries installed on UPPMAX are:
```
CSV
CUDA
MPI
Distributed
IJulia
Plots
PyPlot
DataFrames
```
Installing julia packages

Info

Containers let you install programs without needing to think about the computer environment, like
- operative system
- dependencies (libraries and other programs) with correct versions
Everything is included
Draw-backs
- you install also things that may be already installed
- therefore, probably more disk space is needed

Docker will unfortunately not work on the clusters, since it requires root permission.

However, it is possible to convert a Docker script to a Singularity container.

The UPPMAX staff has already other ways to install most software applications.
Please use Spack only if other ways to install your tool is not possible or very difficult, e.g. requiring very many dependencies and it is not available through, e.g. EasyBuild.
UPPMAX Spack user guide

Unless your script or program is in the active path, you run it by the full path or ./<file> if you are in the present directory.

Keypoints

You have got an overview of the procedures to install packages/libraries and tools on Bianca through the wharf
If you feel uncomfortable or think that many users would benefit from the software, ask the support to install it.