Résumé :

Data platforms are advanced and distributed information systems that manage the entire data lifecycle through ecosystems of services glued and automated by metadata. Creating effective data platforms requires addressing several issues, including (i) choosing appropriate services based on the data processes to support, (ii) finding data representations that are flexible enough to store heterogeneous data and enable efficient queries, (iii) defining querying systems to query data distributed in different engines, and (iv) leveraging meta-data for automated and smart management. Such challenges must be tackled by techniques and methodologies involving different stakeholders (e.g., designers, end users, cloud service providers) and must be tested on real case studies to understand their real power. This is of particular interest following the increasing interest in digital transformation (e.g., the Agritech sector) fueled by the ever-increasing data generated by more and more pervasive systems.