Exploring Architectural Solutions for Implementing the FAIR Principles in Big Data Environments

Abstract

The concept of Open Science has emerged as a major enabler for scientific collaboration. To develop repositories adhering to this concept, the FAIR Principles have been proposed. However, fulfilling these principles can be challenging when dealing with a significant volume, variety, and velocity of data and metadata. A suitable solution is to develop a Software Reference Architecture (SRA) that considers the characteristics of big data environments and the FAIR Principles. Despite the importance of this solution for Open Science, existing literature lacks a big data SRA that achieves full FAIR compliance. In our research, we address this gap by proposing architectural solutions for the implementation of FAIR-compliant big data sharing repositories. We validate these solutions through case studies and performance evaluations. Future contributions include developing algorithms to instantiate the proposed architectures and creating FAIR-compliant artificial datasets to assist in further validations.