Towards FAIR and federated data ecosystems for interdisciplinary research

Area of Science:

• Computer Science
• Information Science
• Research Infrastructure

Background:

• Scientific data management faces challenges from exponential data growth and fragmentation.
• Traditional centralized systems struggle with data volume and hinder cross-domain collaboration and reproducibility.
• Concerns regarding data sovereignty and governance are increasing.

Purpose of the Study:

• Propose FAIR and federated Data Ecosystems as an improved architectural pattern for research data.
• Integrate decentralized systems into research infrastructure to promote collaboration.
• Address fragmentation and reproducibility issues in scientific data management.

Main Methods:

• Introduce a layered architecture (governance, data, service, application) based on Data Commons, Data Meshes, and Data Spaces.
• Focus on adaptive metadata management, simplified user interaction, robust security, and transparent data transactions.
• Implement a decentralized peer-to-peer network supporting compute-to-data and data-to-compute paradigms.

Main Results:

• Preserve domain-specific expertise and control while facilitating data integration.
• Enable horizontal scaling of research infrastructure.
• Support interoperability requirements and domain autonomy.

Conclusions:

• FAIR and federated Data Ecosystems provide a practical path towards integrated research infrastructure.
• This model enables researchers to build on existing work while maintaining control over data and resources.
• Enhances scientific reproducibility and cross-domain collaboration.