Sustained involvement across SESAME, MYRTE, PATRICIA, and PASCAL — all focused on liquid-metal-cooled reactor safety, thermal hydraulics, and transmutation research.
CENTRO DI RICERCA, SVILUPPO E STUDI SUPERIORI IN SARDEGNA SOCIETÀ A RESPONSABILITÀ LIMITATA
Sardinian HPC research centre applying computational simulation, deep learning, and data infrastructure across nuclear safety, biomedicine, and 3D visualization.
Their core work
CRS4 is a multidisciplinary research centre in Sardinia, Italy, specializing in high-performance computing, advanced visualization, and data-intensive science. They apply computational methods across diverse domains — from nuclear reactor safety simulations and energy system optimization to biomedical data analysis and 3D capture technologies. Their core strength lies in building the software and computational infrastructure that makes complex scientific data usable, whether that means deep learning on heterogeneous hardware, cloud-based research data platforms, or immersive volumetric video systems.
What they specialise in
DeepHealth (their largest single grant at EUR 832,500) combined deep learning with HPC for biomedical applications; V-HDR-V applied similar compute-intensive methods to volumetric video.
EVOCATION, Scan4Reco, and V-HDR-V span 3D scanning of cultural heritage, advanced geometric computing, and volumetric HDR video capture and rendering.
Third-party roles in EJP RD, EOSC-Life, and EOSC Future indicate they provide computational and data platform services to European research infrastructure initiatives.
NETFFICIENT addressed integrated storage and ICT tools for smart energy communities; SENDER focuses on consumer engagement and demand response.
PhenoMeNal built e-infrastructure for metabolic phenotyping analysis; EJP RD focuses on FAIR data and omics for rare diseases.
How they've shifted over time
In their early H2020 period (2015–2018), CRS4 spread across energy storage tools, nuclear safety simulation, metabolic phenotyping, cultural heritage scanning, and 3D visualization — reflecting a broad computational services profile. From 2019 onward, their work sharpened around two poles: deep learning and cloud-based research data infrastructure (DeepHealth, EOSC-Life, EOSC Future) and continued nuclear safety work (PATRICIA, PASCAL). The shift shows a centre moving from general-purpose HPC applications toward AI-driven computation and European-scale data platforms.
CRS4 is converging on AI/HPC infrastructure for science, making them a strong partner for projects needing serious computational muscle behind data-heavy research questions.
How they like to work
CRS4 overwhelmingly operates as a consortium partner (11 of 15 projects) or third-party contributor (3 projects), with only one coordinator role (V-HDR-V, a focused MSCA fellowship). With 435 unique partners across 41 countries, they function as a widely connected specialist that teams invite for computational and data expertise. This pattern signals an organization that is easy to integrate into large consortia and brings technical depth without competing for leadership.
CRS4 has collaborated with 435 unique partners across 41 countries, giving them one of the broader networks for a mid-sized Italian research centre. Their projects span pan-European consortia in nuclear safety, EOSC infrastructure, and health research, with no single geographic concentration beyond the natural EU-wide spread.
What sets them apart
CRS4 occupies a rare niche: a single research centre that credibly operates in both nuclear reactor simulation and biomedical deep learning, connected by their underlying HPC and data engineering capabilities. Unlike university labs tied to one discipline, CRS4 functions as a computational problem-solving service that moves between domains. For consortium builders, this means one partner who can handle the heavy computing and data infrastructure work regardless of the application sector.
Highlights from their portfolio
- DeepHealthLargest single grant (EUR 832,500) combining deep learning with HPC for biomedical applications — represents their peak funding and core technical identity.
- V-HDR-VTheir only coordinator role, an MSCA fellowship for volumetric HDR video — shows where they chose to lead and invest their own research agenda.
- PATRICIAPart of the MYRRHA ecosystem for nuclear waste partitioning and transmutation — demonstrates sustained commitment to advanced nuclear safety over nearly a decade across multiple projects.