About
We will present the Rede de Armazenamento Seguro (RAS), the Secure Storage Network of RNP (Brazil's National Research and Education Network), as the core pillar of a sovereign, scalable, and high-performance data ecosystem in Brazil. Our initial strategy leverages an "On-Premise as a Service" (OPaaS) approach through a strategic OPEX partnership, allowing rapid deployment with low capital expenditure (low-CAPEX) in National Data Centers (CNDs) of TIER III standard. This business model is crucial for delivering S3-compatible Storage at a predictable, transparent, and significantly more accessible cost than global hyperscalers, solving the vendor lock-in challenge for the academic community. Furthermore, we will highlight our 2026 expansion roadmap toward a fully open-source approach, including the Ceph platform over Open Compute Project (OCP) hardware, aiming for complete control over the value chain, increased resilience, and the consolidation of national expertise in open infrastructure.
From a technical and scalability perspective, the primary differentiator of RAS is its native integration with the RNP's "Rede Ipê" and e-Science network. This high-performance connectivity (Tb/s backbone) is a non-negotiable requirement for modern science, enabling the efficient, large-scale transfer of petabytes of data generated by global HPC/AI and research projects. The presentation will focus on how the S3 API compliance of RAS facilitates integration with scientific data workflows, and our next steps toward federated interoperability, which directly aligns with the standards and APIs of the CS3 and OCM communities. The system's technical robustness, supported by backends like Ceph, meets the high durability and security requirements needed for critical research data.
By sharing the RAS business strategy, technical roadmap, and governance model, our goal is to offer the CS3 community a replicable blueprint for a collective and federated data infrastructure in the Global…
The talk
Credits and licence
- Institution
- CERN
- Speaker
- Mateus Rodrigues Oliveira
- Year
- 2026
- Language
- English
- Subjects
- Computer science
- Source
- CERN
- Licence
- CERN, as CERN states it. CERN's terms
- Attribution
- © CERN. CERN provides its audiovisual media free of charge for educational and informational use, credited to CERN; this lecture is shown in CERN's own player.
- Who can watch
- CERN gives no rating. Invisible Network shows it to Tweens profiles and older, judged from what the source says about it. How courses are placed