HPC @ Uni.lu

High Performance Computing in Luxembourg

2018 Software Environment

The UL HPC platform provides a large variety of scientific applications to its user community.
They are both domain-specific codes and general purpose development tools enabling research and innovation work in a variety of computational fields.

A new software environment is now available on the Iris and Gaia clusters, with software packages updated to their recent 2018 editions. It currently includes 198 applications and libraries, and while a detailed list is available as always at hpc.uni.lu/users/software, we can highlight:

  • Machine Learning: PyTorch, TensorFlow, Keras, Apache Spark
  • Math & Optimization: MATLAB, Mathematica, CPLEX
  • Physics & Chemistry: GROMACS, ESPResSo, QuantumESPRESSO, Meep, ABINIT, NAMD, NWChem, VASP, CRYSTAL
  • Bioinformatics: SAMtools, BEDTools, BWA, BioPerl, FastQC, PLINK, SNPTEST, FASTX-Toolkit, TopHat, Bowtie2, Trinity, BLAST+, ABySS, mpiBLAST, HTSlib
  • Computer Aided Design & Engineering, Computational Fluid Dynamics: ANSYS, OpenFOAM
  • Development & Performance: GNU Compiler Collection (C/C++/Fortran), Intel Parallel Studio (C/C++/Fortran/MKL, Advisor, ITAC, VTune Amplifier), PGI Compilers (C/C++/Fortan), Python, R, Perl, Go, Rust, Allinea Forge & Performance Reports, Scalasca & Score-P, PAPI
  • Container systems: Singularity

One of our focus points since last year has been to enable Deep Learning and Data Analytics work on the HPC platform through the addition of dedicated high-profile community tools such as PyTorch and TensorFlow. We note that the commercial packages MATLAB and Mathematica also bring highly specialized features for AI/ML/DL and analytics in their recent versions. Their capabilities will be further enhanced by the additions of GPUs and big memory systems to the Iris cluster later this year.

Many more such tools exist and can be delivered also through the use of Containers, with Singularity HPC-oriented container system now being globally available and coming with Docker (and Docker registry) compatibility.

The new software environment will become the default starting with Monday, September 17th, and the current (2017-era) environment will then become legacy.
Any new additions and updates to software packages for 2018-2019 will happen in the 2018 environment.

New environment modules have been created which allow switching between the yearly software set releases on the HPC platform, they can be found and detailed using module avail swenv and module show $modulename (e.g. module show swenv/default-env/v1.1-20180716-production).

The new software sets can be used immediately by including before the usual module load command:

(Iris/Gaia): module load swenv/default-env
(Iris/Gaia): module load swenv/bioinfo-env

This will load the latest, 2018 software set swenv/default-env/v1.1-20180716-production which can be explored using module avail and any specific application profiles loaded as usual with module load.

Users that require applications from the 2017-era software set after September 17th will have to explicitly load the specific module(s) for it:

(Iris): module load swenv/default-env/v0.1-20170602-production
(Iris): module load swenv/bioinfo-env/v0.1-20170602-production

(Gaia): module load swenv/default-env/v0.3-20170713-production
(Gaia): module load swenv/bioinfo-env/v0.3-20170713-production