Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Preliminary findings on establishing a privately governed data ecosystem for MLOps data sharing

Malysh, Konstantin LU orcid ; Runeson, Per LU orcid and Linåker, Johan LU orcid (2025) p.480-486
Abstract
The integration of machine learning (ML) into software operations (MLOps) has significantly increased the demand for data to train and test models, making data sharing and reuse essential to mitigate the high costs of data collection and preprocessing. This study aims to advance the understanding of data sharing within an MLOps context by examining the establishment and governance of a data ecosystem, specifically focusing on the WARA-Ops initiative.Through a case study approach, this research investigates the technical platform, processes and governance of WARA-Ops data ecosystem by conducting interviews with ecosystem members, as well as observational studies of the technical infrastructure.Six interviews were held, with the platform... (More)
The integration of machine learning (ML) into software operations (MLOps) has significantly increased the demand for data to train and test models, making data sharing and reuse essential to mitigate the high costs of data collection and preprocessing. This study aims to advance the understanding of data sharing within an MLOps context by examining the establishment and governance of a data ecosystem, specifically focusing on the WARA-Ops initiative.Through a case study approach, this research investigates the technical platform, processes and governance of WARA-Ops data ecosystem by conducting interviews with ecosystem members, as well as observational studies of the technical infrastructure.Six interviews were held, with the platform orchestrators and representatives of four organizations interested in participating, as well as representatives of a different data sharing platform for validation purposes. Observations on the platform and surrounding ecosystem structure, governance and challenges were made, specifically centered around privacy, integrity and trust between the actors.As the WARA-Ops platform is still in the early development stage, it is important for all the parties to align their interest and expectations in order to create a safe and sustainable Open Data Ecosystem. (Less)
Please use this url to cite or link to this publication:
author
; and
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
host publication
Proceedings : 2025 IEEE 45th International Conference on Distributed Computing Systems Workshops: ICDCSW 2025 - 2025 IEEE 45th International Conference on Distributed Computing Systems Workshops: ICDCSW 2025
pages
480 - 486
publisher
IEEE
DOI
10.1109/ICDCSW63273.2025.00095
project
Next Generation Communication and Computational Infrastructures and Applications (NextG2Com)
B2B Data Sharing for Industry 4.0 Machine Learning
WASP: Wallenberg AI, Autonomous Systems and Software Program at Lund University
language
English
LU publication?
yes
id
cfa7fdad-24a8-49a3-8059-e40890772789
date added to LUP
2025-12-09 10:51:48
date last changed
2025-12-12 10:20:57
@inproceedings{cfa7fdad-24a8-49a3-8059-e40890772789,
  abstract     = {{The integration of machine learning (ML) into software operations (MLOps) has significantly increased the demand for data to train and test models, making data sharing and reuse essential to mitigate the high costs of data collection and preprocessing. This study aims to advance the understanding of data sharing within an MLOps context by examining the establishment and governance of a data ecosystem, specifically focusing on the WARA-Ops initiative.Through a case study approach, this research investigates the technical platform, processes and governance of WARA-Ops data ecosystem by conducting interviews with ecosystem members, as well as observational studies of the technical infrastructure.Six interviews were held, with the platform orchestrators and representatives of four organizations interested in participating, as well as representatives of a different data sharing platform for validation purposes. Observations on the platform and surrounding ecosystem structure, governance and challenges were made, specifically centered around privacy, integrity and trust between the actors.As the WARA-Ops platform is still in the early development stage, it is important for all the parties to align their interest and expectations in order to create a safe and sustainable Open Data Ecosystem.}},
  author       = {{Malysh, Konstantin and Runeson, Per and Linåker, Johan}},
  booktitle    = {{Proceedings : 2025 IEEE 45th International Conference on Distributed Computing Systems Workshops: ICDCSW 2025}},
  language     = {{eng}},
  month        = {{07}},
  pages        = {{480--486}},
  publisher    = {{IEEE}},
  title        = {{Preliminary findings on establishing a privately governed data ecosystem for MLOps data sharing}},
  url          = {{http://dx.doi.org/10.1109/ICDCSW63273.2025.00095}},
  doi          = {{10.1109/ICDCSW63273.2025.00095}},
  year         = {{2025}},
}