Shipping your analysis: Open containerization tooling for web-users bringing on-demand computation
Shipping your analysis: Open containerization tooling for web-users bringing on-demand computation


This is a discussion of technical concerns and considerations in building a fully open source, on-demand, container-based geodata science platform for the web, using Python, Laravel/PHP, PostGIS and Kubernetes. This is particularly aimed at use-cases where at least some users triggering analysis are non-technical consumers.

In 2016, the OurRagingPlanet project was a joint winner of Northern Ireland's first open data challenge to build an educational tool using local, and external, open data, funded by the Departments of Economy and Finance, through the OpenDataNI initiative. The platform itself, developed through a local open source business group, launched in April 2017.

OurRagingPlanet illustrates outcomes of imagined natural disasters on Northern Ireland's landscape. This is targeted to secondary school students, using open data to contextualize human impact, by simulating damage to familiar local landmarks and risk estimations across regional landscape. The Python-based simulations draw on open data to let teachers generate a basic lava flow or storm surge centred on dynamically chosen locations, and to lead their students through the evolution of an emergency.

However, this stack was designed to be sector agnostic and, more generally, it provides a platform to run Python-based data-based simulations on demand, geographical and otherwise. All components of the OurRagingPlanet stack are available under open source licenses (including MIT and AGPL) to facilitate expansion of applications, and help bridge the gap between simulation and web.

Data is maintained in a PostGIS database, retrieved by a Laravel-based central control server and passed for rendering in the Leaflet/VueJS Javascript client. When a specific simulation is requested, the physical and geographic information about the request, along with the necessary open datasets, is passed to the Python simulation server through a Redis datastore.

The simulation server uses a range of Python data analysis tools to generate transient field and point results, which are returned to the control server for later Leaflet layer (or textual) display. Python analysis tools that will be discussed include pandas, FEniCS and geographical libraries.

Drawing on internal developer experience of creating such systems, the project is moving towards sandboxed Python simulation, where scientific web-users can upload safely contained scripts, manipulating data and return their own simulation output. Modelling and technical challenges, and options, in implementing such web systems will be discussed."


Phil Weir is a web developer and mathematician, particularly focused on open source consultancy. He has worked both in academia, modelling floating structures in the Antarctic, and industry, simulating problems in bio-engineering.

He runs Flax & Teal, a Belfast-based open source consultancy, coordinating and supporting web and scientific projects. He is heavily involved in the Belfast development industry, as a co-organizer of the Belfast Linux User Group and Northern Ireland Developer Conference. His language experience is primarily in Python, PHP, C/C++ and Fortran.

Privacy Settings
We use cookies to enhance your experience while using our website. If you are using our Services via a browser you can restrict, block or remove cookies through your web browser settings. We also use content and scripts from third parties that may use tracking technologies. You can selectively provide your consent below to allow such third party embeds. For complete information about the cookies we use, data we collect and how we process them, please check our Privacy Policy
Consent to display content from - Youtube
Consent to display content from - Vimeo
Google Maps
Consent to display content from - Google