Scientific Computing

OCTOPUS – The hybrid GPU supercomputer octopus is the flagship machine, composed of 136 Nvidia GeForce GTX Titan X accelerators. The design of the octopus was realised together with Colfax. The discovery cluster is a 3 nodes GPU/MIC hybrid machine, and served to test different hardware accelerators and intra-node memory bandwidth optimisation.

Link: octopus homepage

Computation resources to the UNIL community (DCSR)

The Scientific Computing and Research Support Unit  (Division calcul et soutien à la recherche, DCSR, in French) provide computation and storage resources to the UNIL community as well as transverse expertise to the various faculties of the University and its affiliated institutes. The expertise ranges from high performance computing (HPC) programming support to full stack development consulting missions.

Link: DCSR homepage

–> back to the list


UNIL GeoCatalog (accessible to UNIL members only): Geospatial database for GIS operations (i.e., shapefile and raster data) are stored of a GIS server. Data are freely available for UNIL  Staff and students. Information includes:  topographic maps, digital terrain models (altitudes), aerial photographs, Satellite photographs, Hydrographic networks, Road networks, Land cover, census data (population, and buildings), etc..

Federal Office of Topography (Swisstopo): The Federal Office of Topography “measures” Switzerland. It surveys and documents the landscape and the underground, and produces high-quality spatially-referenced geodata. Its most important products include landscape and height models, aerial images, orthophotos, geological data and maps, reference data and of course the well-known series of national maps.

Swiss Open Government data: is the Swiss public administration’s central portal for open government data. A centralised and reliable tool, the website offers easy access to the public data of the Confederation, cantons and communes.

OpenStreetMap: OSM is a collaborative project to create a free editable geographic database of the world. The geodata underlying the maps is considered the primary output of the project.[2] (accessible to UNIL members only, ask to Daily Earth Data to See Change and Make Better Decisions. Planet provides daily satellite data that helps businesses, governments, researchers, and journalists understand the physical world and take action.

Global Land Cover Maps: annual coverages from 1992 to 2020, about 300 me resolution. Freely available, simply subscribe and download.

Landsat Data Access: Landsat Level-1 data, as well as Level-2 and Level-3 science products held in the USGS archive, have been available for download at no charge from a variety of data portals. This page provides information about searching and downloading Landsat data and science products.

Copernicus Open Access Hub (previously known as Sentinels Scientific Data Hub) provides complete, free and open access to Sentinel-1, Sentinel-2, Sentinel-3 and Sentinel-5P user products, starting from the In-Orbit Commissioning Review (IOCR).

Copernicus European Ground Motion Service provides consistent and reliable information regarding natural and anthropogenic ground motion over the Copernicus Participating States and across national borders, with millimetre accuracy.

AlpArray is a broad cooperation program around the Alpine orogen in the geosciences. The main field effort was the establishment and operation of a state of the art seismological network (details), ALL the data is now openly and freely available (details). Similarly, a the first pan-Alpine gravity dataset is now available (details).

RaspberryShake produces low-cost seismometers, accessible to anyone from professionals to schools and hobby users. The global network has over 1500 stations running. The waveform images are available real-time, the actual waveforms can be downloaded with 30 minutes delay. Station view, Data centre, and more available from

–> back to the list

Learning resources

Getting Started with Machine Learning – a list of resources to help environmental scientists get started with machine learning. It included tutorials with codes, literature reviews, textbooks (often including code), and datasets that are already pre-processed for ML applications.

Hands-on introduction to Multiple Point Statistics – introduction to stochastic simulation using MPS, a modelling approach based on the use of training images with the aim of generating realistic heterogeneity characterizing natural processes.

Solving PDEs in parallel on GPUs with Julia – ETH’s course 101-0250-00L on solving partial differential equations (PDEs) in parallel on graphical processing units (GPUs) with the Julia programming language.

–> back to the list

Open source programming languages

Python is a programming language that lets you work more quickly and integrate your systems more effectively.

R is a free software environment for statistical computing and graphics.

If you are a Data Scientist, chances are that you program in either Python or R!

Julia was designed from the beginning for high performance.

Julia brings a refreshening programming mindset to the Data Science community

Read more: R vs. Python vs. Julia (Published in Towards Data Science)

–> back to the list

Web services for developers

Git is a software that helps developers and data scientists to store and manage their codes and other text-based resources, as well as track and control changes and collaborate with others.
Several web-based services host resources managed with Git and help to make accessible these resources. GitHub is the best-known website, others are Gitlab , Bitbucket, or Codeberg .

Jupyter is a Free software, open standards, and web services for interactive computing across all programming languages.

Google Colaboratory, or “Colab” for short, is a hosted Jupyter notebook service that requires no setup to use, while providing access free of charge to computing resources including GPUs.

–> back to the list