{"resourceId":"storagereview-netapp-novus-pnfs-ai-factory-20261001","versions":[{"version":"external-352c4832605f0f8870305c84b01eb3ec771bbdc9faa29727d60dc2a2e3d2a3fc","resource":{"id":"storagereview-netapp-novus-pnfs-ai-factory-20261001","title":"StorageReview: NetApp Novus 100 TB/s AI-factory storage claim rests on Omdia modeling of a smaller audited test","organization":"StorageReview (trade press) reporting on NetApp","sector":"Strategic partner: AI data infrastructure","geography":"Global (announced at NetApp INSIGHT 2026, Las Vegas)","publishedAt":"October 1, 2026","publicationDate":"2026-10-01","eventDate":"2026-09-29","sourceName":"StorageReview","sourceLabel":"NetApp Novus Splits pNFS Metadata From AFF A90 Data Nodes for a 100 TB/s AI Factory Target, Orderable Now","sourceUrl":"https://www.storagereview.com/news/netapp-novus-pnfs-metadata-aff-a90-data-nodes-100-tb-s-ai-factory","evidenceClass":"independent-reporting","outcomeClass":"emerging","topics":["infrastructure","data-security","operating-model"],"finding":"StorageReview reports NetApp Novus separates pNFS metadata (Novus Data Director on qualified Supermicro servers) from AFF A90 data nodes running ONTAP, uses stock NFSv4.2 pNFS clients, and is orderable now, with lighthouse customers expected within a month or two of INSIGHT. The article stresses that the 100 TB/s figure is a projection: Omdia audited near-linear scaling on a smaller test and modeled the larger result. It also notes AIDE advanced features remain preview with no GA date, and that NetApp's planned EEA sovereign Keystone support uses the same in-region passport-holder model NetApp uses for its U.S. public sector offering.","sledRelevance":"Limited: Novus targets roughly a hundred AI-factory operators with tens of thousands of GPUs, far beyond typical state, local or K-12 scale. Possible relevance to large research universities and state research computing consortia; most SLED buyers would see AFX or AIPod-class systems instead.","evidence":"Independent trade reporting based on vendor briefing and release; no independent benchmark was run by StorageReview. Omdia test report, configuration and raw results are not linked.","architectureImplications":"Standards-based pNFS client avoids proprietary agents; requires a lossless Ethernet client fabric separate from the GPU backend fabric; initial release is hardware-bound to AFF A90 variant and qualified servers.","governanceImplications":"Buyers should treat headline throughput as a modeled target and require workload-specific acceptance tests in contracts.","securityPrivacyImplications":"Retains ONTAP data services, multi-tenancy and cyber-resilience per vendor; sovereign support models are relevant for regulated data but EEA-specific here.","caveats":"Vendor-sourced figures relayed by trade press; projection not measurement; scale far exceeds most SLED environments.","streamIds":["netapp"],"roles":{"sales":"Sales framing: Novus is not a mainstream SLED offer. Mention it only to large research universities or consortia planning GPU clusters at national-lab scale, and frame 100 TB/s as a modeled design target, not a measured result. For most agencies and campuses the relevant takeaway is that NetApp is investing in standards-based pNFS and metadata disaggregation, which may later reach smaller systems. Do not imply a public-sector reference customer exists.","engineering":"Engineering framing: the architecture uses stock in-kernel NFSv4.2 pNFS Flex Files clients with nconnect and GPUDirect Storage, a separate Data Director metadata tier, and AFF A90 data nodes on a lossless Ethernet client fabric. For research computing evaluations, request the Omdia audit details, test configuration and mixed-workload results, because the article confirms the headline figure is extrapolated. Compare against existing parallel file systems on small-file metadata load and checkpoint bursts.","delivery":"Delivery framing: any research-computing engagement should begin with a workload characterization and a bounded proof of concept on the first hardware-bound configuration. Plan for network fabric work, since the client fabric must be lossless and separate from GPU backend traffic. Timelines are uncertain because lighthouse customers were only expected within a couple of months of announcement. Write acceptance tests around the institution's own throughput and metadata needs rather than vendor maxima."},"roleDetails":{"sales":{"customerProblem":"Large research institutions need storage that keeps GPU clusters fed without proprietary clients.","stakeholders":["Research computing director","HPC architect","CIO","Sponsored research office"],"discoveryQuestions":["What GPU count and throughput do your planned workloads actually need?","Is a proprietary parallel file system client a problem today?","What metadata scale (file counts) do you expect?"],"valueHypothesis":"For very large GPU clusters, a standards-based pNFS design may reduce client management overhead, pending independent validation.","potentialEngagement":"Workload characterization and benchmark planning session for a research computing team.","unsupportedClaims":["100 TB/s as a measured result","Relevance to typical agency or K-12 workloads","Existing public-sector reference deployments"]},"engineering":{"fit":"Only very large research GPU clusters.","architectureIntegration":"Novus Data Director metadata tier on qualified servers plus AFF A90 data nodes; stock pNFS clients on GPU servers over a separate lossless Ethernet fabric.","prerequisites":["Lossless Ethernet client fabric","Linux clients with NFSv4.2 pNFS Flex Files","Qualified hardware"],"constraints":["Initial release hardware-bound","Scale targets are modeled","Early availability timing"],"security":"Vendor states ONTAP data services, multi-tenancy and cyber-resilience are preserved; verify tenant isolation in shared research clusters.","proofOfValue":["Obtain Omdia audit methodology","Run mixed small-file and checkpoint workloads","Measure metadata operations at expected file counts"]},"delivery":{"work":"Bounded proof of concept and fabric readiness assessment.","dependencies":["Network team for lossless fabric","Hardware lead times","NetApp lighthouse program"],"ownership":"Research computing / HPC operations.","skillsAdoption":"Storage and HPC staff need pNFS and fabric tuning skills.","governanceCheckpoints":["Benchmark plan approval","Fabric design review","Go/no-go after proof of concept"],"acceptanceCriteria":["Proposed: sustained throughput meets institution-defined target on its workloads","Proposed: metadata latency within agreed bound at target file count","Proposed: no client-side proprietary software required"],"risks":["Overestimating need relative to scale","Early-release maturity","Network fabric cost"]}},"retrievedAt":"2026-10-30T03:04:55Z","enrichedAt":"2026-10-30T03:12:55Z","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Interpretation: staff skills in storage, identity and security operations are the main workforce dependency; no accessibility findings in the source.","procurementImplications":"Interpretation: buy against documented, testable requirements and acceptance tests rather than vendor maxima.","operatingModelImplications":"Interpretation: ongoing operations sit with central or departmental IT storage teams; no change in operating model is reported by the source.","sourceVerification":{"openedUrl":"https://www.storagereview.com/news/netapp-novus-pnfs-metadata-aff-a90-data-nodes-100-tb-s-ai-factory","referenceExcerpt":"The scaling figures are projections built on a smaller audited test.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}