Research Infrastructures (RIs) serve as the backbone of modern scientific exploration, generating vast and complex datasets that drive innovation across multiple disciplines. With advancements in sensors, cameras, and digital technologies, RIs have become big data factories, producing and publishing ever-growing datasets in FAIR data repositories. However, many scientific communities struggle to fully exploit those datasets due to the lack of technical expertise, scalable computational resources, and ready-to-use environments. Without proper tools for data staging, analysis, and AI-driven insights, the potential of these datasets remains untapped.

RI-SCALE is a newly funded EC project designed to bridge this gap by co-design, prototyping, and validating the Data Exploitation Platform (DEP). This powerful new RI technology will enhance scalable data accessibility and analysis. The DEP will provide research communities with integrated environments that bring together scientific data, AI tools, and high-performance computing, enabling deeper insights and groundbreaking discoveries. Built upon open-source components, open standards, and APIs, the DEP will ensure interoperability across various data holdings and computational environments. Initially, the platform will support environmental and health science RIs, with plans to expand across the broader RI landscape through strategic partnerships between scientific data providers, compute service providers, and AI experts.

The RI-SCALE project brings together four thematic RIs – EISCAT (Atmospheric Physics), ENES (Climate), BBMRI (Biobanking), and Euro-BioImaging (Imaging Resources), two compute providers from EGI – Technical University of Wien and TUBITAK, and three Data Spaces – Copernicus, DestinE, and EUCAIM.

In this article, we will examine two examples of how RIs, specifically EISCAT and ENES, will benefit from the DEP technology. These examples highlight the significant impact the Data Exploitation Platform will have on scalable data accessibility and analysis.

EISCAT operates radar systems that focus on the ionosphere and space weather, gathering valuable data on Earth's upper atmosphere and geospace. EISCAT_3D, an advanced radar facility currently under construction, will provide detailed, volumetric data of the upper atmosphere, including three-dimensional atmospheric dynamics. This facility will generate vast amounts of data, ranging from hundreds of terabytes to several petabytes annually.

However, gaining insights from this ever-growing data poses significant challenges, including limited compute resources and difficulties in sharing data effectively with the global research community. By expanding  EISCAT’s infrastructure with DEP, these challenges can be overcome.

The Data Lifecycle Management capability of DEP will allow EISCAT’s datasets to be replicated on-demand across scalable e-infrastructure resources. This will enable EISCAT to store large datasets remotely, making it easier to access and share data with collaborators worldwide. Furthermore, the DEP's cloud-based solutions will enhance AI model deployment, enabling EISCAT to use advanced machine learning techniques to process and analyse complex data more efficiently. This AI-driven approach could significantly improve real-time space weather prediction capabilities, or issue warnings to space satellite operators concerning dangerous space debris.

ENES supports a large scientific community focused on climate modeling, particularly through the Coupled Model Intercomparison Project (CMIP), which provides crucial climate projections that inform global climate policy and research. CMIP6, the most recent iteration of the project, generates over 20 petabytes of data from global climate simulations. As the volume of climate-related data continues to grow, ENES faces challenges in providing researchers with efficient and scalable access to these datasets.

RI-SCALE’s DEP will address these challenges by offering a scalable solution for managing and processing CMIP data. By integrating DEP into ENES’s data infrastructure, researchers will benefit from an integrated, cloud-based platform that supports data orchestration, data transfer, and AI model development. The platform will allow researchers to analyse climate data more efficiently, leveraging AI frameworks that can accelerate data processing and enhance the quality of climate models.

The interoperability of DEP with various data formats and computational resources will enable ENES to tap into new computational power, such as national cloud systems, high-performance computing (HPC) resources, and external EuroHPC allocations. This will ensure that ENES can process large-scale climate simulations, enhancing both the precision and scalability of climate predictions.

Moreover, the AI frameworks integrated into DEP will facilitate the creation of new models and algorithms that can be used to improve the accuracy of climate simulations. AI tools within the platform will also help enhance the FAIR (Findable, Accessible, Interoperable, Reusable) nature of ENES data, promoting greater accessibility and usability for the broader research community.

Although EISCAT and ENES serve different aspects of atmospheric science—space weather and climate modeling, respectively—both RIs face similar challenges in managing and exploiting large-scale scientific data. The integration of DEP into both infrastructures will enable them to share data more effectively, improve the quality of their data, and integrate data across disciplines. For instance, the AI models developed for analysing space weather data from EISCAT could be adapted to enhance climate modeling in ENES, fostering cross-disciplinary insights.

By leveraging DEP’s modularity and extensibility, both EISCAT and ENES will be able to build on each other’s successes. The interoperability between these RIs will support a cross-disciplinary ecosystem that drives new research innovations and supports the broader scientific community.

The DEP is set to become an essential tool for a broad spectrum of data provider research infrastructures across Europe, expanding them with computational service offerings towards their users. With its focus on scalability, open-source principles, and AI-driven tools, DEPs will facilitate data valorisation, enabling RIs to maximise the value of their scientific data. Through the collaborative development of this platform, RI-SCALE will tackle a wide range of scientific and technical challenges. While initially benefiting EISCAT, ENES, BBMRI, and Euro-BioImaging, in future, the platform’s impact will extend across multiple research domains, providing scalable solutions for a wide range of RIs.

As DEP continues to evolve, it will support the sustainability of research infrastructures by providing energy-efficient solutions for AI model training and data processing, making it possible to handle even more complex datasets. With a focus on ensuring the long-term viability of these platforms, DEP will ensure that RIs can continue to scale and support research across multiple disciplines, through exploitational platforms jointly provisioned by RIs and e-infrastructures.