Published November 25, 2025 | Version v1
Presentation Open

EXTRACT-TASKA: A Distributed Workflow Orchestrator for Radio Astronomy Data Processing

  • 1. Observatoire de Paris, Université de Recherche Paris Sciences et Lettres
  • 2. ROR icon Laboratoire d'études spatiales et d'instrumentation en astrophysique
  • 3. ROR icon Université Paris Cité
  • 4. ROR icon Observatoire de Paris
  • 5. Ekinox
  • 6. PSL universite
  • 7. CNRS Délégation Paris B
  • 8. Barcelona Supercomputing Center

Description

We present the distributed workflow orchestrator framework developed within the EXTRACT project, for the TASKA (Transients Astrophysics with and SKA pathfinder). The orchestrator is built on existing software developed and maintained  by the partner of the EXTRACT project. We demonstrate that a simple workflow for radio astronomy interferometric imaging can be run with this framework, using various cloud infrastructures. It has been tested with a private cloud provider (OVH), as well as on academic cloud infrastructures (EGI/CESNET, EOSC EU Node, ObsParis local OKD cluster). The demonstration can run from a jupyter notebook (e.g., on the EOSC EU Node), and the processing is run remotely on the selected cloud infrastructure. 

Notes (En)

The presentation was given to the OAEG4 group on Nov. 25th 2025.  

The attached videos show:

  • demo-eosc-eu-node.mp4: a demo running on the EOSC EU Node Jupyter Notebook service, with the actual data processing running on OVH. 
  • demo-okd-eosc.mp4: a demo running on a local laptop, with the actual data processing running on the EOSC EU Node cloud computing service (OKD instance). 

The EXTRACT project has received funding from the European Union’s Horizon Europe programme under grant agreement number 101093110.

Files

extract-cecconi.pdf

Files (307.8 MB)

Name Size Download all
md5:c8c27adea2af7e1e88df6b7c473baa25
16.5 MB Preview Download
md5:c9e300e22e87faee16fc3494b02e5293
269.4 MB Preview Download
md5:95cc703cd32bc500d59189482820b8b6
22.0 MB Preview Download

Additional details

Funding

European Commission
EXTRACT - A distributed data-mining software platform for extreme data across the compute continuum 101093110