{"resourceId":"bristol-brics-genci-federated-learning-isambard-jean-zay-adastra-20261021","versions":[{"version":"external-610d1e0cdc4df420a7a6bd5894dd4d31926222252cf11e93fbcb1218f19c96ff","resource":{"id":"bristol-brics-genci-federated-learning-isambard-jean-zay-adastra-20261021","title":"UK and France plan federated learning across national AI supercomputers (Isambard-AI, Jean Zay, Adastra) so sensitive data stays in-country; announced at GTC Berlin with NVIDIA technical support","organization":"Bristol Centre for Supercomputing (University of Bristol); GENCI (France); IDRIS; CINES; University of Cambridge; NVIDIA","sector":"Public research computing (national AI supercomputers)","geography":"United Kingdom and France","publishedAt":"2026-10-21","publicationDate":"2026-10-21","eventDate":"2026-10-21","sourceName":"University of Bristol, Bristol Centre for Supercomputing","sourceLabel":"University news release","sourceUrl":"https://www.bristol.ac.uk/research/centres/bristol-supercomputing/articles/2026/httpswwwbristolacukresearchcentresbristol-supercomputingarticles2026federated-learning-brics-genci.html","evidenceClass":"vendor-claim","outcomeClass":"emerging","topics":["infrastructure","data-security","governance-procurement","operating-model"],"finding":"BriCS and GENCI announced a federated learning plan under the UK-France ConcordAI partnership to connect Isambard-AI (NVIDIA Grace Hopper, part of the government-funded AI Research Resource) with France's Jean Zay (NVIDIA) and Adastra (other hardware), so researchers in both countries can train larger models while each keeps control of its own machines and data. Each site trains locally and shares only model updates, so sensitive data in fields such as healthcare, defence or aerospace need not leave its country. Teams at Bristol, Cambridge, IDRIS and CINES, with NVIDIA's support using the open-source NVIDIA FLARE, have been testing with increasing model sizes and checking results match single-machine training; direct connection is targeted by the end of 2026. The release also notes Juelich and GENCI interest in linking JUPITER AI Factory and AI France Factory infrastructure, and GENCI plans to connect its supercomputers with French hyperscalers for startups and industry.","sledRelevance":"Directly relevant to public research computing: two publicly funded national systems are sharing capacity across borders without moving data, including across mixed vendor hardware. The model transfers to U.S. state or university consortia (for example, multi-institution health research) that face data-residency or data-use restrictions, with the caveat that it is a plan with testing under way, not a production service.","evidence":"Joint announcement by the public operators, framed as a plan. It reports testing activity but no results, model sizes, performance overheads or governance agreements. NVIDIA's role is technical support and software.","architectureImplications":"Federated training over heterogeneous hardware (NVIDIA and non-NVIDIA) using NVIDIA FLARE; only model updates cross sites.","governanceImplications":"Each country retains control of machines and data; cross-border collaboration rests on a bilateral partnership rather than data-sharing agreements for raw data.","securityPrivacyImplications":"Keeps raw sensitive data in-country; the release does not address risks from shared model updates (such as inference about training data).","caveats":"Announcement of a plan; no published test results, overhead figures or allocation rules; completion target end of 2026.","streamIds":["nvidia"],"roles":{"sales":"Sales framing: public research centres can pool AI compute across jurisdictions without moving sensitive data, and can do it across mixed hardware. Interpretation: for U.S. multi-state or multi-university research consortia, federated learning is a discussion about data-use restrictions and shared capacity, not a hardware sale. The release reports a plan and tests without results, so engagements should be scoped as feasibility studies.","engineering":"Engineering framing: federated training with NVIDIA FLARE across sites, including non-NVIDIA hardware, with equivalence checks against single-machine training. Pre-sales engineers should assess network links between sites, software stack consistency, model-update security and how schedulers coordinate jobs. The source gives no overhead or accuracy figures, so validation must measure both before claiming parity. Secure aggregation options and update encryption also need review, since the release does not describe them.","delivery":"Delivery framing: federated research needs agreements on data governance, scheduling windows, software versions and incident handling across institutions. Delivery teams should start with a non-sensitive model, scale size gradually as BriCS and GENCI did, and record equivalence results. Each site keeps its own operations, so a joint coordination group is needed. Clear escalation paths help. Acceptance criteria below are proposals, not reported outcomes."},"roleDetails":{"sales":{"customerProblem":"Researchers need more compute than one centre holds but cannot move sensitive data across jurisdictions.","stakeholders":["Research computing directors at each site","Data governance or IRB officers","National or state research funders"],"discoveryQuestions":["Which datasets cannot leave their institution or state?","Do partner sites run different hardware?","What network capacity links the sites?"],"valueHypothesis":"Interpretation: federated training could expand usable compute for restricted data; no performance data yet.","potentialEngagement":"Federated learning feasibility study across two or more research computing centres.","unsupportedClaims":["Parity with centralised training (not yet reported)","Production readiness by a specific date","Elimination of all privacy risk"]},"engineering":{"fit":"Multi-site research with data-residency constraints.","architectureIntegration":"NVIDIA FLARE coordinating training across national supercomputers; local data, shared model updates.","prerequisites":["Inter-site connectivity","Aligned software stacks","Agreed model and evaluation protocol"],"constraints":["Heterogeneous hardware","Separate schedulers and allocation policies","Target completion end of 2026"],"security":"Secure aggregation and protection of model updates should be assessed; not described in the source.","proofOfValue":["Compare federated and single-site model accuracy","Measure training overhead from coordination","Test with progressively larger models"]},"delivery":{"work":"Set up cross-site coordination, software alignment, joint testing and governance agreements.","dependencies":["Bilateral or consortium agreement","Network links","FLARE deployment at each site"],"ownership":"Each centre for its system; joint steering group for the federation.","skillsAdoption":"Research software engineers need federated learning skills at each site.","governanceCheckpoints":["Data governance approval per dataset","Equivalence test sign-off","Security review of update exchange"],"acceptanceCriteria":["Proposed: federated model within an agreed accuracy margin of centralised baseline","Proposed: no raw data transfer confirmed by audit","Proposed: coordination overhead reported per run"],"risks":["Scheduling conflicts between centres","Software drift across sites","Model-update leakage of sensitive information"]}},"retrievedAt":"2026-10-22T03:00:08Z","enrichedAt":"2026-10-22T03:10:04Z","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Aims to widen access to larger compute pools for researchers and startups in both countries.","procurementImplications":"Demonstrates interoperability across vendors via open-source software, which may reduce lock-in for federated use cases.","operatingModelImplications":"Two national operators coordinate scheduling and software stacks for joint training while retaining separate operations.","sourceVerification":{"openedUrl":"https://www.bristol.ac.uk/research/centres/bristol-supercomputing/articles/2026/httpswwwbristolacukresearchcentresbristol-supercomputingarticles2026federated-learning-brics-genci.html","referenceExcerpt":"only the 'lessons learned' are shared between sites, instead of copying vast amounts of data between countries.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}