NetApp is a storage and data-management vendor whose authority concentrates in one of the most mature governed data foundations in the market (Layer 1A) and the storage-adjacent infrastructure around it: a software-defined storage fabric (0C), best-in-class data movement (Layer 1C), and a hybrid-multicloud data control plane (Layer 2A). For the AI execution, reasoning, and application layers, NetApp is the data half of a meet-in-the-middle AI factory (AIPod): the compute, model-serving runtime, retrieval, and applications are NVIDIA's, sold alongside through the channel, not NetApp's.
The capture is decoupled and Oracle-shaped: open at the surface, captive beneath. Data is reached through standard protocols (NFS, SMB, S3) and travels freely: uniquely, as first-party services on all three clouds (Amazon FSx for NetApp ONTAP, Azure NetApp Files, Google Cloud NetApp Volumes). But the value accumulates in ONTAP's opinions (SnapMirror relationships, clone hierarchies, efficiency and snapshot policies, the data-management workflows enterprises build on), and those do not lift. Leaving ONTAP is a multi-year rebuild even though the bytes move freely; running it on a hyperscaler does not loosen the hold, because it is still ONTAP. Seven of eleven scored components are Ceded, nearly all in the data plane where the ONTAP opinions live.
NetApp's push up the stack, the AI Data Engine, has been reset. NetApp's documentation promoted its metadata catalog, vectorization, and retrieval endpoints to GA in 2026, and the instrument scored them; the GA documentation was withdrawn in July, and the September 29 Insight release repositions AIDE as a metadata service, with semantic search, knowledge graphs, and agentic activation as capabilities NetApp is building toward. The cells follow the documentation: 1A reads strong on Data Classification, which infers sensitivity and personal-data types per file and makes them queryable, while the AI-metadata catalog stays in private preview, and 1B reads moderate on the AIPod Mini Enterprise RAG validated design, the retrieval NetApp documents today. The mature data foundation plus data movement remains the deployable core.
The meet-in-the-middle structure is the defining authority finding. Unlike Dell and Cisco (primes that sell, curate, and support the full NVIDIA stack as their own AI factory, and are therefore credited for the runtime and retrieval they deliver), NetApp brings the storage half of AIPod while NVIDIA brings the compute, runtime, retrieval, and application ecosystem, and the channel integrates. Each vendor is credited for its half: NetApp owns the data plane; NVIDIA owns the intelligence. The NVIDIA dependency is heavy exactly where NetApp partners (compute, Layer 2B runtime, Layer 1B retrieval, Layer 3 applications) and near-empty where NetApp owns the layer (1A, 1C, 2A).
The buyer's trade: the buyer gets the most operationally proven, most portable, best-governed enterprise data foundation available (the safest home for data across on-prem and every major cloud), and in exchange accumulates ONTAP-specific opinions that are a multi-year lift to leave, even as the bytes stay open. For AI specifically, NetApp anchors the data; the intelligence is bought alongside from NVIDIA. NetApp's bet is that the governed data foundation is the durable position in the AI stack, and that when the AI Data Engine ships broadly it can convert that foundation into an AI-native data platform. The sharp contrast is VAST, its closest competitor: the same data-foundation strength, but where VAST verticalized into native retrieval and an agent runtime, NetApp partners for them while its own AI-native layer matures.
Layer-by-layer status: Layer 0A (No Compute on NetApp's Paper), Layer 0B (ONTAP Cluster and Storage Switches on NetApp Part Numbers), Layer 0C (ONTAP on AFF, ASA and Disaggregated AFX, With StorageGRID for Object), Layer 1A (Data Classification With Queryable Sensitivity Per File; AI-Metadata Catalog in Private Preview), Layer 1B (No Retrieval on NetApp Paper; AIPod Mini Is a Reference Design), Layer 1C (Movement and Tiering), Layer 2A (Data-Infrastructure Control Plane; No GPU Scheduling), Layer 2B (Runtime is NVIDIA's (Meet-in-the-Middle)), Layer 2C (No Reasoning Plane (Data Guardrails ≠ Agent Governance)), Layer 3 (+1) (No Value Plane: Data Foundation Beneath Others' Apps).
Assessment framework: 4+1 Layer AI Infrastructure Model. Scoring model: Decision Authority Placement Model (DAPM) — Retained, Delegated, or Ceded. Published by The CTO Advisor LLC (DBA The Advisor Bench). Author: Keith Townsend. Date assessed: October 27, 2026. Version: v1.27 - 4+1 v3: Layer 0 Sub-Layers.
NetApp is a storage and data-management vendor whose authority concentrates in one of the most mature governed data foundations in the market (Layer 1A) and the storage-adjacent infrastructure around it: a software-defined storage fabric (0C), best-in-class data movement (Layer 1C), and a hybrid-multicloud data control plane (Layer 2A). For the AI execution, reasoning, and application layers, NetApp is the data half of a meet-in-the-middle AI factory (AIPod): the compute, model-serving runtime, retrieval, and applications are NVIDIA's, sold alongside through the channel, not NetApp's.
The capture is decoupled and Oracle-shaped: open at the surface, captive beneath. Data is reached through standard protocols (NFS, SMB, S3) and travels freely: uniquely, as first-party services on all three clouds (Amazon FSx for NetApp ONTAP, Azure NetApp Files, Google Cloud NetApp Volumes). But the value accumulates in ONTAP's opinions (SnapMirror relationships, clone hierarchies, efficiency and snapshot policies, the data-management workflows enterprises build on), and those do not lift. Leaving ONTAP is a multi-year rebuild even though the bytes move freely; running it on a hyperscaler does not loosen the hold, because it is still ONTAP. Seven of eleven scored components are Ceded, nearly all in the data plane where the ONTAP opinions live.
NetApp's push up the stack, the AI Data Engine, has been reset. NetApp's documentation promoted its metadata catalog, vectorization, and retrieval endpoints to GA in 2026, and the instrument scored them; the GA documentation was withdrawn in July, and the September 29 Insight release repositions AIDE as a metadata service, with semantic search, knowledge graphs, and agentic activation as capabilities NetApp is building toward. The cells follow the documentation: 1A reads strong on Data Classification, which infers sensitivity and personal-data types per file and makes them queryable, while the AI-metadata catalog stays in private preview, and 1B reads moderate on the AIPod Mini Enterprise RAG validated design, the retrieval NetApp documents today. The mature data foundation plus data movement remains the deployable core.
The meet-in-the-middle structure is the defining authority finding. Unlike Dell and Cisco (primes that sell, curate, and support the full NVIDIA stack as their own AI factory, and are therefore credited for the runtime and retrieval they deliver), NetApp brings the storage half of AIPod while NVIDIA brings the compute, runtime, retrieval, and application ecosystem, and the channel integrates. Each vendor is credited for its half: NetApp owns the data plane; NVIDIA owns the intelligence. The NVIDIA dependency is heavy exactly where NetApp partners (compute, Layer 2B runtime, Layer 1B retrieval, Layer 3 applications) and near-empty where NetApp owns the layer (1A, 1C, 2A).
The buyer's trade: the buyer gets the most operationally proven, most portable, best-governed enterprise data foundation available (the safest home for data across on-prem and every major cloud), and in exchange accumulates ONTAP-specific opinions that are a multi-year lift to leave, even as the bytes stay open. For AI specifically, NetApp anchors the data; the intelligence is bought alongside from NVIDIA. NetApp's bet is that the governed data foundation is the durable position in the AI stack, and that when the AI Data Engine ships broadly it can convert that foundation into an AI-native data platform. The sharp contrast is VAST, its closest competitor: the same data-foundation strength, but where VAST verticalized into native retrieval and an agent runtime, NetApp partners for them while its own AI-native layer matures.
CPUs, accelerators, servers, integrated systems, OS and hypervisor, physical plant
All AI compute in NetApp's AI factories is NVIDIA's: AIPod pairs ONTAP storage with DGX systems. AIPod Mini is the exception, Intel Xeon 6 with Intel's open-source OPEA RAG framework.
The DX50 data-compute node ships with an L4 for storage-side data-engine acceleration, not general AI compute.
NetApp sells no compute. The AI compute is NVIDIA's and Cisco's: AIPod pairs NetApp storage with NVIDIA DGX, and FlexPod AI pairs AFF with Cisco UCS and NVIDIA. Both are validated, channel-assembled designs where NetApp brings the storage half. AIPod Mini's Intel Xeon 6 compute isn't NetApp's paper either. Calibration: WEKA, DDN, VAST and Everpure read gap here too; Dell and HPE sell storage and compute together.
Nothing offered and nothing inherited from NetApp at 0A: Absent.
The DX50 data-compute node carries an NVIDIA L4 for storage-side data-engine acceleration, named in the NVIDIA note, not as a chip. Roadmap, not GA: NVIDIA RTX PRO Servers (Blackwell) support.
Interconnects, switching and routing, network virtualization
Cluster interconnect for ONTAP clusters past two nodes, and the AFX switch network that connects controller nodes to NVMe-oF storage shelves over RoCE; NetApp reference configuration files
The GPU networking in AIPod is NVIDIA's, on NVIDIA's paper. NetApp's own switches are Cisco Nexus.
NetApp sells the switches its storage needs and no others. Cisco Nexus 9336C-FX2, 9332D-GX2B, 9364D-GX2A and 9808 switches ship on NetApp part numbers with NetApp reference configuration files. They carry the cluster interconnect for ONTAP clusters past two nodes, and the AFX switch network that connects controller nodes to NVMe-oF storage shelves over RoCE. The buyer gets a storage back end supported as one system. The architect's concern is reach. It's a storage back-end interconnect only, not an AI fabric: the GPU networking in AIPod is NVIDIA's. Calibration: VAST reads moderate on the same shape; WEKA, DDN and Everpure read gap; Dell reads strong on an AI fabric it sells.
Ceded where the grade rests. NX-OS makes the routing calls on a configuration NetApp defines, with no documented per-decision override: vendor decides, visible, not overridable, Ceded. Portability: Cisco Nexus switches on NetApp part numbers read Ceded to Cisco under the channel-substitution and proprietary-networking rules.
Arrays, storage servers and software, front-end protocols: where the data sits
NetApp's unified all-flash systems running ONTAP, serving block and file from one system: the AFF A90 and C80 key specifications list CIFS, FC, iSCSI, NFS v3 through v4.2, NFS over RDMA, NVMe/FC, NVMe/TCP and S3. The protocols are their own chip. The architecture and ONTAP's opinions cannot be lifted to another vendor as deployed. Proprietary NetApp platform: opinions captive, no open exit.
Standard fabric interfaces connecting compute to the storage estate: NVMe-over-Fabrics, NVMe/TCP, Ethernet, pNFS/RDMA. Substitutable switching against multi-vendor standards; the connectivity opinions lift. Mirrors how VAST's NVMe-oF fabric is scored.
Multiprotocol data-management OS (NFS, SMB, S3, iSCSI, NVMe-oF) with SnapMirror, FlexClone, efficiency, and tiering, on AFF, ASA and AFX and as Cloud Volumes ONTAP, which the enterprise runs on instances in AWS, Azure or Google Cloud. Standard protocols at the surface, which are their own chip, but the ONTAP data-management opinions are captive: leaving ONTAP is a multi-year rebuild, and running it on a hyperscaler doesn't change that. The clouds' own ONTAP services are the clouds' chips, because they sell and run them; NetApp doesn't. Proprietary NetApp platform: opinions captive, no open exit.
Geo-distributed S3-compatible object storage. The S3 API is its own chip. The grid behind it, its information lifecycle management policies, geo-distribution and erasure-coding schemes, is NetApp's and doesn't lift to another S3 store: Ceded.
Autonomous Ransomware Protection's detection and automatic snapshots on ONTAP: protection of the bytes. Ceded, as the chip reads.
NetApp's SAN-only all-flash systems on ONTAP: the ASA A1K key specifications list FC, iSCSI, NVMe/FC and NVMe/TCP. The protocols are their own chip. Proprietary NetApp platform: Ceded.
AFX disaggregates performance from capacity (AFX 1K controllers + NX224 NVMe enclosures, scale-out, DGX SuperPOD-certified). "Clients access the nodes through multiple protocols, including NFS, SMB, and S3," and "configuration options for SAN as well as aggregates and RAID groups have been removed." Proprietary NetApp platform: Ceded.
The S3 API, which StorageGRID implements with documented limitations. Workloads confined to the supported common S3 operations move by changing endpoints, credentials and provider-specific configuration; the data, policies and unsupported service features migrate or adapt. Delegated. The grid and its lifecycle policies are the StorageGRID chip, Ceded.
NFS, SMB, S3, FC, iSCSI, NVMe/FC and NVMe/TCP, the protocols NetApp's key specifications list for AFF and ASA and its documentation lists for AFX; availability varies by system and ONTAP version, and ASA serves block only and AFX file and object only. Independent implementations share them: hosts and applications re-point to another server or target speaking the same protocols, and the data and provider-specific policies migrate with it. Delegated. ONTAP's data management stays in the ONTAP and system chips, Ceded.
AIPod holds the NVIDIA-Certified Storage designation, and AFF A90 and AFX are certified for DGX SuperPOD. ONTAP itself has no NVIDIA dependency.
NetApp owns a software-defined storage architecture. AFX disaggregates ONTAP, separating performance from capacity (AFX 1K controllers and NX224 NVMe enclosures, scaling to 128 controllers and beyond an exabyte) over NVMe-oF, and it's DGX SuperPOD-certified. ONTAP is a multiprotocol data-management OS (NFS, SMB, S3, iSCSI, NVMe-oF) with SnapMirror, FlexClone, efficiency and FabricPool tiering, on AFF, ASA and AFX and as Cloud Volumes ONTAP. StorageGRID adds S3 object, and Autonomous Ransomware Protection adds ML-based detection with automatic snapshots. The buyer gets the most operationally proven data home in the market. The architect's concern is that the value accumulates in ONTAP's opinions: replication topology, clone hierarchies, snapshot and efficiency policies. Leaving ONTAP is a multi-year rebuild. Calibration: WEKA, DDN and VAST read strong on the same function.
Delegated where the grade rests. ONTAP places, tiers and protects the bytes, and a volume move relocates a FlexVol to an aggregate the administrator chooses, nondisruptively, the override rule's reassign: vendor decides, visible, overridable, Delegated. Portability: AFF, ASA, AFX, ONTAP and Autonomous Ransomware Protection read Ceded; StorageGRID's S3 surface and the storage networking read Delegated.
Named, not scored here: Amazon FSx for NetApp ONTAP, Azure NetApp Files and Google Cloud NetApp Volumes are sold and run by the clouds and scored on their rows, Ceded to NetApp on portability with the cloud deciding at runtime (October 21, 2026). Channel note: ONTAP also ships on Lenovo paper (ThinkSystem DM/DG), scored Ceded to NetApp on the Lenovo row. Roadmap, not GA: STX (BlueField-4, Vera Rubin) support.
The data itself, organized and governed: the Governance Catalog that Layer 2C queries
PII/sensitive-data classification and mapping (formerly Cloud Data Sense), observability with AI-driven anomaly detection (formerly Cloud Insights), and ML-based ransomware protection in ONTAP (on by default on recent AFF/ASA platforms). Compliance and resilience governance: proprietary opinions captive to NetApp's tooling, not an AI-metadata catalog. Proprietary NetApp platform: opinions captive, no open exit.
ONTAP, StorageGRID, Data Classification, Insights, and Autonomous Ransomware Protection are NetApp IP over standard protocols. No NVIDIA. (The AI Data Engine's NIM-powered vectorization, now GA and scored at Layer 1B, adds an NVIDIA dependency inside the retrieval surface.)
With ONTAP at 0C, NetApp’s 1A is Data Classification. It reads file content and assigns a category, a sensitivity level, and the personal and sensitive personal data types it finds, and the Investigation page and REST API filter on each of them per file, next to tags the enterprise assigns. That answers what the data is at the object level, over data on NetApp storage. ONTAP and StorageGRID hold the bytes and Autonomous Ransomware Protection protects them, 0C functions, and Data Infrastructure Insights is observability at 2A. The architect’s concern is what the classification covers. It’s compliance-shaped: sensitivity and personal data, not the AI metadata a reasoning plane would add. The AI Data Engine (AIDE) Metadata Engine, the piece that would add it, is back in private preview. 1A is graded on a catalog that classifies the data, and as ruled October 16, 2026, compliance classification counts. Strong. Calibration: VAST reads strong on a catalog of tags the enterprise applies; Nutanix reads moderate on metadata and permissions without content classification; WEKA reads moderate on filesystem fields.
Data Classification reads the content and assigns the classification, and the docs show no way to correct an inferred category or sensitivity level: vendor decides, visible, not overridable, Ceded. Portability: the classifiers and their findings are NetApp’s alone, Ceded.
Named, not scored (October 5, 2026): the AI Data Engine metadata catalog, documented live as AIDE 1.1.0 in private preview. Actions and policies for saved queries are in preview and unscored.
Low-latency retrieval for RAG — vector/hybrid search, context windows
AIDE's vectorization pipeline is NVIDIA-NIM-powered: the embedding intelligence inside NetApp's retrieval surface is NVIDIA's, the same split as every storage vendor's GA retrieval.
In the meet-in-the-middle AIPod, the co-sold retrieval path remains NVIDIA's NeMo Retriever / NIM within NVIDIA AI Enterprise. AIDE now gives NetApp a first-party alternative on AFX estates.
NetApp sells no retrieval product. Its documented retrieval path is the AIPod Mini Enterprise RAG validated design: Intel AI for Enterprise RAG and the OPEA stack on AFF A20 storage and Xeon 6 servers, an air-gapped RAG pipeline over enterprise documents, maintained in NetApp's solution documentation (version 2.0, February 2026). By NetApp's own documentation it's a validated reference design, not a NetApp-paper product: the compute is Supermicro, the switch is Arista, the retrieval software is open-source OPEA cloned from GitHub, and delivery runs through distributors and integrators (the 2B cell records the same reading). A reference design isn't the vendor's paper and earns no grade, so the layer is a gap and AIPod Mini is named. The native retrieval surface the cell once rested on, AI Data Engine's vectorization and RAG endpoints, is no longer documented: the GA pages return 404, and NetApp's September 29, 2026 Insight release repositions AI Data Engine as a metadata service building toward semantic search. Calibration: Dell and Nutanix read moderate on retrieval they sell; VAST reads strong on native vector search.
None on NetApp's paper. An enterprise that builds the AIPod Mini design runs open-source OPEA it operates itself; NetApp validates the stack and sells the AFF storage beneath it.
Named, not scored (October 5, 2026): AI Data Engine vectorization, data collections, retrieval endpoints, and Data Guardrails. Scored GA July 12, 2026 on AIDE 9.18.1 documentation that now returns 404; NetApp's September 29, 2026 Insight release repositions AI Data Engine as a metadata service and lists semantic search, knowledge graphs, and agentic activation as capabilities it is building toward, with no date, so it lives here rather than on the watch-list. Any republished GA documentation returns it to the cell. Named, not scored: NetApp AIPod Mini Enterprise RAG validated design (reference design, not NetApp paper; October 19, 2026 /reconcile).
Move/transform data — ETL/ELT, lineage, cost-aware movement, KV cache tiering
Async and sync replication across sites and clouds: the logistics layer for staging and distributing datasets, including AI training and inference data. ONTAP-to-ONTAP; the replication relationships and opinions do not lift to another platform. Proprietary NetApp platform: opinions captive, no open exit.
FlexCache caches hot data near compute; FabricPool auto-tiers cold data to object/cloud by temperature. A captive ONTAP caching and tiering engine, cost-aware data movement whose opinions are proprietary. Proprietary NetApp platform: opinions captive, no open exit.
Moves and synchronizes data from external, file, and SaaS sources into the estate across hybrid multicloud. NetApp-proprietary sync and migration service; the movement opinions are captive. Proprietary NetApp platform: opinions captive, no open exit.
SnapMirror, FlexCache, FabricPool, and Cloud Sync are NetApp IP. No NVIDIA. (The AI Data Engine's NIM-powered vectorization and Data Sync are now GA. See notes on the pending 1C re-read.)
NetApp's data movement is best-in-class for all data types, and it is a legitimate Layer 1C capability even though it is not marketed for AI: the layer's purpose explicitly includes movement, cost-aware movement, and cache tiering, and AI data logistics are exactly that. SnapMirror replicates and stages datasets across sites and clouds; FlexCache caches hot data near compute; FabricPool tiers cold data to object/cloud by temperature; BlueXP/Console Copy & Sync moves and synchronizes data from external and SaaS sources. SnapMirror's reach belongs to the architecture, not to one delivery: it replicates between NetApp's own systems, Cloud Volumes ONTAP and the clouds' ONTAP services, which document it themselves (Amazon FSx for NetApp ONTAP supports SnapMirror and FlexCache; Google Cloud NetApp Volumes' external replication uses bidirectional SnapMirror with ONTAP-based systems, GA May 18, 2026). The clouds run those endpoints, and the movement capability is ONTAP's. An architect leverages these for AI today; the absence of an 'AI' label does not remove the capability. What earns the score is the movement-decision capability: configurable replication topologies, cost-aware tiering policies, and caching policy, where the enterprise accumulates opinions (the same opinions that make these Ceded to ONTAP), not transparent byte-transport, which is a Layer 0/1A throughput property, not a Layer 1C decision. What keeps this cell at moderate is the missing general half: ETL/ELT, transformation, and lineage remain the enterprise's, met with its own tooling (Airflow, Spark, Kubeflow). Best-in-class movement without a general pipeline engine is partial capability, moderate, calibrated against VAST's DataEngine (general, event-driven, strong) and Dell's Dataloop (the moderate anchor), above Nutanix (gap, narrow copy-data management). The AI Data Engine ingest pipeline (sync, classify, vectorize) that also sat here was scored GA on July 12, 2026; its documentation has been withdrawn and NetApp's Insight release now positions AIDE as a metadata service, so it's named, not scored, and the grade never rested on it.
Low for movement: SnapMirror, FlexCache, FabricPool, and Cloud Sync are NetApp IP, the data-movement opinions Ceded to NetApp. The AI-pipeline/transform/lineage half: the enterprise brings its own ETL/orchestration (Airflow, Spark, Kubeflow). Settled October 5, 2026: NetApp's AIDE sync, classify, and vectorize pipeline is no longer documented, so nothing of NetApp's closes that half.
Re-read resolved July 12, 2026 under the general-versus-fixed-function criterion in the methodology: AIDE's pipeline pieces were a purpose-built slice; as of October 5, 2026 they're no longer documented and are named, not scored, and the cell holds at moderate on movement. What would move this cell toward strong: arbitrary/insertable transformation stages, destinations beyond AIDE's own surfaces, or lineage tracking: general-pipeline capability, documented.
GPU scheduling, quotas, RBAC, fair-share scheduling, utilization optimization
Unified provisioning, protection, governance, mobility, and health across the on-prem and multicloud storage estate (AFF/ASA/FAS/AFX, StorageGRID, FSx for ONTAP, Azure NetApp Files, Google Cloud NetApp Volumes). Proprietary control plane: orchestration opinions captive to NetApp. Proprietary NetApp platform: opinions captive, no open exit.
Open-source (Apache 2.0), NetApp-maintained CSI driver provisioning NetApp storage to Kubernetes for AI and container workloads. The consumed interface is CSI/Kubernetes, a genuine multi-vendor standard: provisioning opinions live in Kubernetes-standard objects (StorageClasses, PVCs) and survive a swap to another vendor's CSI driver without rebuilding the layer. What the driver connects to is NetApp storage, and that capture is scored where it lives, at 1A. Matches the Dell CSI Operator precedent.
PII/sensitive-data classification and mapping (formerly Cloud Data Sense), observability with AI-driven anomaly detection (formerly Cloud Insights), and ML-based ransomware protection in ONTAP (on by default on recent AFF/ASA platforms). Compliance and resilience governance: proprietary opinions captive to NetApp's tooling, not an AI-metadata catalog. Proprietary NetApp platform: opinions captive, no open exit.
NetApp does no GPU scheduling, quotas, or fair-share. That function sits with NVIDIA (Run:ai / GPU Operator) and the enterprise's Kubernetes: the same universal Layer 2A gap that Dell and VAST carry. NetApp orchestrates the data infrastructure, not the AI compute.
NetApp orchestrates the data infrastructure, not the AI compute. NetApp Console (formerly BlueXP) is a mature, unified control plane for the entire storage and data estate across on-prem and all three clouds: provisioning, protection, governance, mobility, and health from one surface. Trident is NetApp's open-source Kubernetes CSI driver, dynamically provisioning ONTAP/StorageGRID/FSx/ANF storage to containerized and AI workloads. Together they are genuine infrastructure-orchestration capability for the data plane. The AI core of this layer (GPU scheduling, quotas, fair-share) NetApp does not provide; that belongs to the enterprise's Kubernetes and NVIDIA. So Layer 2A sits at moderate, calibrated to VAST and Dell: a storage/data control plane plus a vendor CSI driver, with GPU scheduling absent or borrowed from NVIDIA (universal across these peers). NetApp Console is a genuinely deep multicloud data control plane, comparable in kind to VAST Polaris; the GPU-scheduling gap is the same one Dell and VAST carry, so it does not dock NetApp below them.
NetApp owns the data-infrastructure control plane (Console) and the CSI driver (Trident): Ceded to NetApp. AI-compute orchestration (GPU scheduling, quotas, fair-share) is not NetApp's at all; the enterprise's Kubernetes and NVIDIA own it. Decision authority (October 5, 2026): Console and Trident execute the StorageClasses, backends, and provisioning actions administrators define. Code decides, visible, overridable, Retained, the judgment test (October 4, 2026).
Trident is open source (Apache 2.0, NetApp-maintained), but it is a NetApp-storage-specific CSI driver: the Kubernetes interface is portable, the driver is not. NetApp Console was renamed from BlueXP in 2025. Placement (October 16, 2026): Data Classification + Data Infrastructure Insights + Autonomous Ransomware Protection reads here under the function audit and the hardware-management placement rule.
Model serving, agent execution, inference APIs, distributed inference
The model-serving and inference runtime in a NetApp-anchored AI factory is NVIDIA NIM / AI Enterprise: the runtime half of the meet-in-the-middle AIPod, co-sold and channel-integrated, not a NetApp-delivered or NetApp-supported solution. NetApp supplies the governed data beneath.
In FlexPod AI (Cisco + NetApp), model serving comes from Red Hat OpenShift AI. Again a partner runtime; NetApp provides the storage.
NetApp has no model serving, no inference runtime, and no agent runtime. None. Where inference runs in a NetApp-anchored stack it is NVIDIA NIM / AI Enterprise (AIPod), Red Hat OpenShift AI (FlexPod AI), or open-source OPEA / Intel AI for Enterprise RAG on Intel Xeon 6 AMX (AIPod Mini with Intel): three partner runtimes, none NetApp's. This is the runtime half of the meet-in-the-middle: NetApp brings the storage, NVIDIA brings the runtime, the channel integrates. Crucially, NetApp is not the prime that sells, curates, and supports the NVIDIA runtime as its own factory, unlike Dell (2B moderate, delivered NVIDIA runtime) or Cisco (2B moderate, own security IP over the runtime). So the capability is not credited to NetApp's row; it is NVIDIA's, scored there. The enterprise inherits NVIDIA's (NIM) or Red Hat's (OpenShift AI) execution layer entirely. NetApp is the data foundation beneath it. This sits below every storage and platform peer at Layer 2B: VAST (strong, built AgentEngine), Dell (moderate, delivered NVIDIA runtime), Nutanix (moderate, NAI serving). Because NetApp built no runtime IP and does not deliver the runtime as its own solution.
Total. NetApp provides no runtime. In a NetApp-anchored AI factory the runtime is NVIDIA's (NIM / AI Enterprise) or Red Hat's (OpenShift AI), co-sold via the channel rather than delivered by NetApp. NetApp is the storage beneath someone else's execution layer.
The meet-in-the-middle distinction is load-bearing: Dell and Cisco are primes that deliver and support the NVIDIA stack as their own factory (and are credited at 2B); NetApp is the storage half of a channel-assembled co-sell, so the NVIDIA runtime is credited on NVIDIA's row, not NetApp's. Assessed July 13, 2026 (NetApp AIPod Mini with Intel, solution brief SB-4332 + TR-5010): fails the whose-paper test and does not move this cell. Per NetApp's own documentation it is a 'validated reference design,' not a NetApp-paper product: compute is Supermicro (222HA-TN-OTO-37), switch is Arista, the runtime is open-source OPEA / Intel AI for Enterprise RAG cloned from GitHub, and delivery is through distributors and integration partners (Arrow, TD SYNNEX; Insight, CDW, Presidio, Long View). NetApp brings storage plus the reference architecture; it does not sell or support the runtime on its own paper. It is a second meet-in-the-middle path (Intel/OPEA-CPU alongside NVIDIA-GPU), which strengthens the storage-half finding rather than closing it.
Policy-driven placement and resource coordination — the Autonomy Layer
Neither NetApp nor NVIDIA provides a reasoning plane in this stack. As with Dell (which sells the full NVIDIA AI factory as prime and is still Layer 2C gap), NVIDIA's stack (AI-Q, Dynamo) is routing and scaffolding, not policy-driven placement. Selling or anchoring an NVIDIA factory does not create a 2C capability.
Applying the 'Routing Is Not Reasoning' test: NetApp has no agent governance, no model routing, no policy-driven inference placement, and no multi-agent orchestration. Its only adjacent capability is the AI Data Engine Data Guardrails, which scan, classify, and exclude sensitive data from AI/RAG pipelines: 'guardrails follow the data.' That is data-access governance (a Layer 1A-style claim), not agent-action governance (Intelligence-2C) and not request-time placement (Infrastructure-2C). It is the wrong function for this layer: a verdict GA does not change: the guardrails shipped (AIDE 9.18.1 U1) and the cell still reads gap, because what shipped governs data, not agents. This is a clean gap. The wider point still holds and is the more important one: no vendor on the instrument productizes a reasoning plane. Some of the infrastructure cohort now score moderate at 2C on shipped agent-governance surfaces (VMware on AgentMinder, Nutanix on Agent Gateway), while Dell, VAST, NVIDIA, and CoreWeave remain gaps. NetApp ships neither the governance surface nor the placement engine. The enterprise retains policy-driven placement and agent governance in full. The live-placement gap is universal across the instrument, noted rather than penalized.
Inverted: there is no Layer 2C to borrow. The enterprise retains policy-driven placement and agent governance entirely. NetApp's data guardrails (GA, scored at 1B) govern data, not agents; the meet-in-the-middle NVIDIA stack has no reasoning plane either.
AI Data Engine Data Guardrails reached GA (AIDE 9.18.1 U1) and are scored inside the 1B retrieval component, where they function. At 2C they remain the wrong function: data-access governance, not an agent or reasoning plane; the gap holds. You can't prompt your way to deterministic output: the controls here answer whether an action is legal, never whether the outcome was right. That deterministic outcome-validator is missing instrument-wide, and like live placement it is noted as universal rather than charged to one vendor.
AI-powered business capabilities — business logic, workflow automation
The application ecosystem delivered alongside AIPod is NVIDIA AI Enterprise (NIM microservices, blueprints, and models), NVIDIA's ecosystem, co-sold through the channel, not a NetApp-curated AI-application program. NetApp supplies the governed data beneath the apps.
NetApp ships no first-party AI applications and curates no AI-application ISV program of its own. Its ecosystem is infrastructure and compute (NVIDIA, Cisco, Lenovo, Intel, Red Hat) and channel (distribution and integration), not value-plane ISVs. In the AIPod meet-in-the-middle, the application ecosystem is NVIDIA's (AI Enterprise) plus the customer's own apps; NetApp is the data foundation beneath the value plane, not a provider of it. This sits below the partner-ecosystem vendors at Layer 3: Dell, HPE, and Cisco curate genuine AI-application ISV programs (Dell's OpenAI/Palantir/ServiceNow, HPE Unleash AI's 26+ ISVs), and below VAST (partner, Cosmos Community partner tracks). NetApp has neither a first-party app nor a curated AI-app ISV program, so it does not address the value plane even via partners: gap, not partner. The value plane is entirely the customer's, NVIDIA's, and ISVs'.
The value plane is the customer's plus NVIDIA's plus ISVs': NetApp provides no AI applications and curates no AI-application ISV ecosystem. The enterprise brings its own apps; NetApp supplies the governed data beneath them.
AIPod and FlexPod AI position NetApp storage in AI solutions, but the application logic and ecosystem are NVIDIA AI Enterprise (NIM/blueprints/models) and the customer's, co-sold via the channel, not NetApp-curated.
NetApp INSIGHT 2025 (Oct 14-16, 2025), NVIDIA GTC 2026 (Mar 16-19, 2026), netapp.com / docs.netapp.com (ONTAP, AFF/ASA/AFX, StorageGRID, AIPod, AI Data Engine, NetApp Console/BlueXP, Trident), AWS/Azure/GCP first-party service docs, analyst/press coverage (TechTarget, Blocks & Files, StorageMath), published 4+1 model. v1.1 (instrument reconciliation): clarified that Layer 1C credits a data-movement decision/orchestration capability (policy-driven replication/tiering/caching), not transparent transport throughput, distinguishing NetApp's movement decisions from CoreWeave-style byte-transport acceleration. v1.2: Trident at 2A corrected Ceded→Delegated: the consumed interface is CSI/Kubernetes, a multi-vendor standard (Dell v2.5 CSI Operator precedent), and Trident is Apache 2.0; the v1.1 Ceded call confused what the driver connects to (NetApp storage, captured at 1A) with where provisioning opinions accumulate (Kubernetes-standard objects). 1A note added recording the Lenovo DM/DG channel pairing (capture is NetApp's through any paper, per the July 2026 OEM-channel retraction). v1.3 (same day, /ga-check): AI Data Engine promoted on doc-confirmed GA (docs.netapp.com/us-en/ai-data-engine, AIDE 9.18.1 U0/U1 and 1.0.0 release notes). 1B gap→moderate with AIDE vectorization/RAG retrieval endpoints scored Ceded; Metadata Engine added as a 1A component (Ceded); the watch-list's native-vector-DB question resolved native. Guardrails remain wrong-function at 2C (gap holds). 1C re-read resolved same day under the newly codified general-versus-fixed-function criterion: AIDE's ingest pipeline scored as a Ceded 1C component, cell holds at moderate: the slice ships, the general half of the layer remains the enterprise's. v1.4 (July 13, 2026): NetApp AIPod Mini with Intel assessed against 2B and ruled out under the whose-paper rule (validated reference design on Supermicro compute, open-source OPEA runtime, SI-delivered, not a NetApp-paper product). No cell moved; recorded in 2B gap/notes as a second meet-in-the-middle path that confirms the storage-half finding. /reconcile (September 1, 2026): 2C peer cohort refreshed after the VMware and Nutanix promotions. /reconcile (October 5, 2026): Layer 3 calibration citations updated for the Layer 3 re-reads (Nutanix, VAST, VMware to partner; CoreWeave, NVIDIA, AMD to gap) and the MongoDB BI ruling; no grade or authority change on this row. /reconcile (October 5, 2026): authority visible set from the cell's own text where it settles it: layer2a true. Direction, grades, and components unchanged. /reconcile (October 5, 2026): authority re-read against product docs under the judgment test, override rule, and Layer 3 ruling: layer2a to code / Retained. Grades and components unchanged. /reconcile (October 5, 2026), Keith's rulings: the AI Data Engine GA documentation the July 12 promotion cited now returns 404 (docs.netapp.com ai-data-engine aide-architecture, aide-interfaces, faq-aide; NetAppDocs/ontap-afx commits d36f66c and ac2b0cf, July 9 and 10, 2026), the live AIDE 1.1.0 docs describe a metadata intelligence service in private preview, and NetApp's September 29, 2026 Insight release (news-rel-20260929-706774) positions semantic search as future. 1A strong to moderate on the catalog-classification criterion (catalog not GA). 1B holds moderate, re-sourced from AIDE retrieval to the AIPod Mini Enterprise RAG validated design (docs.netapp.com netapp-solutions-ai ai-minipod, v2.0 February 2026). 1C holds moderate on movement; the AIDE ingest chip named, not scored. Summary updated. /reconcile (October 5, 2026): authority read against product docs: layer0 visible true; layer1b visible true. Grades and components unchanged. /reconcile (October 5, 2026): Layer 1B authority vendor / Ceded to code / Retained: OPEA's retriever executes the enterprise's configuration (judgment test); pipeline tracing documented. Grade unchanged. v1.14 (Layer 0 sub-layers, October 16, 2026): Layer 0 split into 0A Compute, 0B Network Fabric and 0C Storage Substrate by runtime decision (Keith, October 16, 2026); storage left 1A for 0C, and 1A, now Data Foundation & Governance, is graded on the catalog that classifies the data. 1A moderate to strong; 0A gap, 0B moderate, 0C strong; summary edits (summary worksheet, Keith, October 16, 2026). v1.15 (October 19, 2026, /reconcile): 1B moderate to gap, authority Absent; AIPod Mini is a validated reference design, not NetApp paper (partnership is not paper, October 15, 2026), the reading the 2B cell already carried. 2B authority Ceded to Absent: NetApp offers no runtime and absorbs none beneath a service, the DDN, WEKA and Everpure gap reading. (Same pass) Plain-language pass: methodology rule numbers in the prose and audit notes replaced with the reasoning they stood for; no grade, chip or authority reading changed. v1.16 (October 19, 2026, /reconcile voice pass): em-dashes in the prose replaced with periods, colons, commas or parentheses; wording, grades, chips and authority readings unchanged. v1.17 (October 21, 2026, /reconcile): ONTAP on a cloud's paper ruled Ceded to NetApp on portability, with the cloud deciding at runtime (Keith, October 21, 2026), channel substitution applied to the clouds. 0C: the clouds' ONTAP services (Amazon FSx for NetApp ONTAP, Azure NetApp Files, Google Cloud NetApp Volumes) taken out of the ONTAP chip's name and detail, since the clouds sell and run them and they're scored on the clouds' rows; notes say so, and the ruling code is out of the text. 1C: SnapMirror's cross-delivery reach named in the prose, from AWS's FSx for ONTAP guide and Google's NetApp Volumes release notes (read October 21, 2026). No grade, chip reading or authority reading changed. v1.18 (October 23, 2026, /reconcile): 0C: Three product lines, two with different protocol scopes, were one chip., so both stay unclassified on protocol.) Found by the Layer 0 data pass after the sniff read. Grade and cell authority unchanged. v1.19 (October 23, 2026, storage door pass): ONTAP gets a front-end door for AFF, ASA and AFX; StorageGRID splits into an S3 door and the Ceded grid. Grade unchanged. v1.20 (October 24, 2026, /reconcile): 1B calibration updated: VAST reads moderate at 1B, InsightEngine unscored without a product guide. Found in the 1B dependency audit follow-ups. Grade unchanged. v1.21 (October 24, 2026, /reconcile): 1B calibration restored: VAST reads strong at 1B again, InsightEngine GA on VAST's own announcement. Grade unchanged. v1.22 (October 24, 2026, /reconcile): GA evidence recorded on 1 of 1 Layer 1A chips under the GA evidence table version 2. Grade unchanged. v1.23 (October 24, 2026, /reconcile): GA evidence recorded on 3 of 3 Layer 1C chips under the GA evidence table version 2. Grade unchanged. v1.24 (October 24, 2026, /reconcile): GA evidence recorded on 1 of 1 Layer 0B chips under the GA evidence table version 2. Grade unchanged. v1.25 (October 24, 2026, /reconcile): GA evidence recorded on 9 of 9 Layer 0C chips under the GA evidence table version 2. Grade unchanged. v1.26 (October 24, 2026, /reconcile): GA evidence recorded on 3 of 3 Layer 2A chips under the GA evidence table version 2. Grade unchanged. v1.27 (October 27, 2026): each component now carries a durable offering id; nothing else changed.