This workshop is envisioned as an addition to the Blue-Cloud Training Academy, which aims to stimulate the uptake of FAIR data practices in marine science and neighbouring disciplines, through contributions by key actors such as EuroGOOS and IEEE. The workshop also has a strong overlap with the FAIR-IMPACT project regarding development and uptake of FAIR service descriptions (expansion of FAIR software) and methods for assessment. To manage and provide access to a large amount of data, a federated or distributed Data Lake can be a solution but it presents some technical bottlenecks and sustainability constraints. For the federation of services it is crucial to achieve technical and semantic interoperability between the services for providing added value to users. “Just providing API’s” is not sufficient. As a basis each data access service (being a “plain” dataset access service or a more advanced subsetting service) needs a FAIR service description which includes e.g. the expected input, output, processing capacity, data policy (CC-BY), etc. Both Blue-Cloud 2026 and FAIR-EASE run into this complex data lake challenge which could be well supported by FAIR service descriptions. Blue-Cloud 2026 is active in the marine domain developing virtual labs, work benches for Essential Ocean Variables (EOVs) and a Virtual Research Environment (VRE) on top of Blue Data Infrastructure services, and, FAIR-EASE develops similar services on top of data access services in the multi-disciplinary domain. Both projects (in coordination with other projects (e.g.,EuroSciencesGateway) contribute to the implementation of the EOSC interoperability framework.