BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//talks.osgeo.org//foss4g-europe-2026//speaker//9SNWNM
BEGIN:VTIMEZONE
TZID:EET
BEGIN:STANDARD
DTSTART:20001029T050000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:EET
TZOFFSETFROM:+0300
TZOFFSETTO:+0200
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:EEST
TZOFFSETFROM:+0200
TZOFFSETTO:+0300
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-foss4g-europe-2026-VWRHVX@talks.osgeo.org
DTSTART;TZID=EET:20260701T153000
DTEND;TZID=EET:20260701T160000
DESCRIPTION:The remote execution of processing workflows is a common task i
 n many spatial data infrastructures and projects. To increase interoperabi
 lity\, the OGC (Open Geospatial Consortium) published the [API Processes s
 tandard in 2021](https://ogcapi.ogc.org/processes/). Its RESTful design an
 d use of JavaScript Object Notation (JSON) encoding make it suitable for c
 loud environments. [Pygeoapi](https://pygeoapi.io/) is an open-source impl
 ementation of this standard.\n\nIn pygeoapi\, several plugins are availabl
 e and a manager component must be implemented to manage  process jobs. A c
 ommon feature of the built-in managers is that the processing jobs are exe
 cuted directly within the pygeoapi Python environment. Hence\, a job with 
 high resource demands influences the resource requirements and usage of th
 e pygeoapi instance itself. Recognizing the limitations of pygeoapi’s bu
 ilt-in job managers regarding isolation and resource handling\, we develop
 ed the [pygeoapi-K8s-manager](https://github.com/52North/pygeoapi_k8s-mana
 ger).\n\nResource sharing and non-existent job isolation are some of the d
 isadvantages of this architecture. Due to the resource-intensive nature an
 d need for scheduled execution of certain processes within our projects\, 
 we had to run a “heavy” pygeoapi deployment in our cluster.  We also n
 eeded to execute processes in diverse runtime environments outside of Pyth
 on\, e.g.\, using the CUDA Fortran model execution.\n\n We decided to deco
 uple the management and execution layers to address these demands. Having 
 already deployed a pygeoapi instance  in our K8s cluster\, it made sense t
 o take advantage of the cluster's processing capabilities. Our team used K
 8s-CronJobs for process scheduling and K8s-Jobs for execution. The Kuberne
 tes API server handled the process management and pygeoapi provided an int
 erface. The resulting\, generic “pygeoapi-k8s-manager” was developed b
 ased on [EOX IT Services GmbH’s](https://eox.at) [“pygeoapi-kubernetes
 -papermill“](https://github.com/eoxhub-workspaces/pygeoapi-kubernetes-pa
 permill).\n\nBy decoupling management and execution\, we were able to defi
 ne complex process requirements such as using GPUs via Job properties. An 
 autoscaler installed in the cluster applies these properties. This  enable
 s on-demand provision of the requested resources.\n\nOur team implemented 
 two processes: a HelloWorld-K8s process and a process to run generic image
 s. The first process demonstrates how to run a preconfigured image. The ge
 neric process enables image configuration via the pygeoapi configuration f
 ile.\n\nWe will present the current pygeoapi-K8s-manager implementation\, 
 future development plans and illustrate its application through exemplary 
 use cases\, such as data ingestion\, flood modelling and ship voyage optim
 ization workflows. Listeners will gain practical insights into how Kuberne
 tes and OGC API Processes can improve your geospatial data processing work
 flows\, e.g.\, by reducing resource requirements. The talk will cover sust
 ainable resource management and explain the operation of pygeoapi in a clo
 ud-native environment. We aim to encourage wider adoption\, feedback\, and
  contributions to these ongoing developments through this conference.
DTSTAMP:20260604T233805Z
LOCATION:Auditorium
SUMMARY:Optimizing resource usage of interoperable geospatial processing in
 frastructures with Kubernetes - Eike Hinderk Jürrens\, Martin Pontius
URL:https://talks.osgeo.org/foss4g-europe-2026/talk/VWRHVX/
END:VEVENT
END:VCALENDAR