30 oktober | Cloudscaping Geo

De digitalisering van geo-informatie evolueert richting van cloud-native technologieën. Deze ontwikkeling biedt grote kansen: schaalbaarheid, flexibiliteit, snelle innovatie en toegankelijkheid van data en diensten. Maar er zijn ook vragen: over cloud-native standaarden, over cloudbeleid en maatschappelijke verantwoordelijkheid.

Op 30 oktober 2025 namen we samen met experts en ervaringsdeskundigen een deepdive in geo in de cloud. Een onderwerp waar nog veel meer over te ontdekken valt! We horen daarom graag welke thema's verdere aandacht verdienen. En op welke onderwerpen we elkaar verder kunnen helpen. 

Presentations

Workshops Round 1

Workshops Round 2

Plenary session

⌚9.45 
Future Data Infrastructure - Jeremy Tandy

Jeremy is a Principal Fellow at the Met Office, the UK's government agency for weather and climate. He has worked for many years as an engineer and architect in IT, and now focuses on the big challenges to make weather and climate data more useful to the UK, Europe and the World. He leads the World Meteorological Organisation's transition of 190-plus countries to open, Web-based data exchange and has written open standards for W3C and OGC including the W3C/OGC Spatial Data Web Best Practices. 

Presentation
Building on experiences, insight and examples from the weather and climate community, this presentation unpacks the thinking behind putting large-scale data in the Cloud, the implications of doing so, and how the challenges arising can be addressed.

⌚10.30 
Capacity Building for Cloud-Native Geospatial Data Use in the Netherlands - Serkan Girgin

Dr. Serkan Girgin is the founder and director of the Center of Expertise in Big Geodata Science (CRIB), focusing on geospatial big data, cloud computing, and optimizing geocomputing workflows for performance and energy efficiency. He develops tools and platforms to support research software and data management, and has over 30 years of experience in GIS, remote sensing, and large-scale web applications. He is an eScience Center Fellow, and has been recognized as a SURF Research Support Champion in the Netherlands.

Presentation
Accessing and processing large geospatial datasets is becoming increasingly challenging as data volumes grow. The upcoming CLOUD-NES project will empower Dutch research communities and public institutions to work efficiently with cloud-native geospatial data. By providing cloud-optimized datasets, co-located analysis platforms, evidence-based benchmarks, and hands-on training, CLOUD-NES will demonstrate the advantages of modern cloud workflows over traditional methods. This keynote will introduce the project’s community-driven approach, outline its planned solutions, and invite participants to explore ways to expand its impact across domains.

⌚11.30
Lessen uit de loopgraven van de geo IT sector - Rosalie van de Maas

Rosalie van der Maas MSc is co-founder en CEO van Ellipsis Drive. Ellipsis Drive is een cloud native data management oplossing die gebruikers vanuit iedere workflow naadloos toegang geeft tot ieder type geodata. Ze heeft een achtergrond in Aardwetenschappen, Milieu Wetenschappen, Internationale politiek, Internationale ontwikkeling en Innovatie. Voor het starten van haar bedrijf in 2019, werkte Rosalie als programmamaker op het thema Big Data, Technology and Society bij Pakhuis de Zwijger en bij het Ministerie van Buitenlandse Zaken op het consulaat in Chicago. ​

Rosalie neemt je mee in de start van Ellipsis Drive. Welke lessen heeft zij geleerd over de werking van de markt en de rol van de overheid? 

Workshops round 1

⌚13.00 
Workshop: The world in a box

As a psychologist who became a data scientist with a fascination for technology, assets and GIS Sjoerd tries to make smart systems that help humans interact with the world around them, make complex spatial systems manageable, and model our city environment in a human-centric way. In today's Dutch landscape where many of our challenges revolve around dealing with space and how to use that, Sjoerd works on tackling these problems using cutting-edge technologies!

In this workshop you will explore the possibilities of discrete global grid systems (DGGS) and how they can aid in making complex analytics such as AI/ML inferencing on distributed cloud systems very scalable. In a GIS world where more and more information will be gained from unstructured data such as images, pointclouds and meshes, these tools will be invaluable for the future GIS expert!

⌚13.00 
Workshop: GIS en de Cloud: best practises, ontwikkelingen en aandachtspunten

Hoe positioneer je geodata in je eigen huidige GIS-omgeving en de omgeving van nieuwe gebruikersgroepen? Tijdens deze sessie zullen Marcel de Rink (namens Esri in Regiegroep Geostandaarden) en Boris Minnaert (Esri Architect) ingaan op cloud ontwikkelingen die betrekking hebben op Geo. Diverse ervaringen die Esri heeft bij klanten zullen worden gedeeld. De focus ligt op waar de mogelijkheden en aandachtspunten liggen. Marcel en Boris kijken hierbij naar zowel de datalaag, applicatielaag als ontsluitingswereld in Cloud omgevingen en de mogelijke impact op standaarden. 

⌚13.00 
Workshop: Cloud-optimized vector formats: a deep dive into GeoParquet

Joris Van den Bossche is a core contributor to Pandas and Apache Arrow, one of the maintainers of GeoPandas and Shapely, and involved in starting the GeoParquet format. He did a PhD at Ghent University and VITO in air quality research, worked at the Paris-Saclay Center for Data Science and at Voltron Data contributing to Apache Arrow. Currently he is a software engineer at Fused and a freelance open source software developer and teacher.

In the context of cloud native geospatial workflows, the usage of standardized file formats that are optimized for usage in the cloud, is a crucial aspect. Especially when data grows larger, it is important to have efficient access to (subsets of) your data. On the vector side of geospatial, GeoParquet is one of the available formats. GeoParquet defines how to store geospatial vector data in Apache Parquet, which is a widely used columnar storage format.
This workshop will give a brief overview of what makes file formats cloud optimized and the different options for vector data, and then dive deeper specifically into GeoParquet, illustrating it with practical examples using Python.

Workshops round 2

⌚14.00
“Zijn we er al?”

Edwin van Eersel en Pieter Koppelaar zijn allebei Solution Architect bij de gemeente Rotterdam op het Data en Analyse domein. Samen beschikken zij over ervaring op datascience, datamanagement, reporting, ETL, infrastructuur, interfacing & security. 

Deze sessie is een reisverslag van de gemeente Rotterdam over de realisatie van een modern datalandschap in de cloud. “Are we there yet?” refereert naar verschillende momenten in die reis waarin we jullie mee willen nemen. Het is een verhaal over geitenpaadjes, steile paadjes, over zon, wind en regen en lastige paspoortcontroles. Kortom: wat ging goed, wat ging fout, wat hebben we geleerd, waar staan we nu en wat komt er nog op ons af. En wat zijn jullie reisplannen?​

⌚14.00
Overview of activities at TU Delft related to (3D) cloud-optimised formats

Hugo Ledoux is an associate-professor in 3D geoinformation at the Delft University of Technology in the Netherlands. He holds a PhD in computer science from the University of South Wales (UK) and a BSc in geomatics engineering from the Université Laval (Québec City, Canada). For his research, he is particularly interested in combining the fields of GIS and computational geometry. He is currently working on the efficient storage of 3D city models with CityJSON, the reconstruction of massive TINs, the validation and the automatic repair of polygons and polyhedra as found in GIS, and the semantic classification of urban textured meshes. He is also the programme director of the MSc Geomatics at TUDelft.

In this technical session, we will first introduce a new course in the MSc Geomatics program, scheduled for this fall. The course will provide an introduction to various cloud-optimised geospatial formats. We will then present our efforts to develop cloud-optimised formats for 3D city models. We have already developed FlatCityBuf, a new cloud-optimised format based on FlatBuffers and CityJSON. We will discuss its pros and cons and demonstrate how it could be used to disseminate the 3DBAG.

Finally, we plan to explore Parquet as an additional format for 3D city models. We have drafted initial ideas on how to encode CityJSON into Parquet, but we are particularly interested in gathering feedback from the community: How are others using Parquet for 3D geospatial data? What kinds of analytical queries are you running? What tools and libraries are you using to read and write Parquet files? During the session, we aim to present our ideas and collect suggestions and insights from participants.

⌚14:00
Introduction to Zarr

Francesco Nattino is a Senior Research Software Engineer at the Netherlands eScience Center, the Dutch national center of expertise for software in academic research. In his position, he collaborates with researchers on topics related to Environment and Sustainability, combining experience in data handling and HPC with a more recently developed passion for geospatial applications.

Zarr is an open and versatile file format designed to store large multidimensional array data. It works efficiently in the cloud, and, while not being domain specific, it has found broad applications in the context of geospatial datasets. In this workshop, I will introduce the Zarr format and discuss its strengths and limitations with respect to other array/raster data formats (e.g. NetCDF, COG). I will show how existing datasets in legacy file formats can be offered as if they were Zarr via virtualization techniques. Finally, I will present current best practices for usage together with SpatioTemporal Asset Catalogs (STACs).
 

⌚15:15 
Paneldiscussion lead by Bart de Lathouwer (Geonovum) 

Bart de Lathouwer interviews our presenters Jeremy Tandy (UK MET Office), Rosalie van der Maas (Co-founder & CEO at Ellipsis Drive) and Serkan Girgin (TU Twente), and participants to round up.

 

> Meld je aan

Geen updates meer missen?

Automatisch op de hoogte blijven? Meld je aan voor één van onze nieuwsbrieven.