Home EnterpriseNetApp AFX AI Portfolio Turns Enterprise Data Into AI Fuel

NetApp AFX AI Portfolio Turns Enterprise Data Into AI Fuel

by Harold Fritts

NetApp launches AFX AI portfolio to deliver disaggregated storage in high-performance environments. And it comes with NetApp AI Data Engine.

As AI shifts from a research project to an enterprise operation, customers require robust, integrated storage solutions that minimize risk, reduce complexity, and lower costs. Enterprise AI projects need solutions that ensure security, reliability, and operational efficiency to achieve a strong return on investment.

NetApp AFX hardware stack

The NetApp AFX AI portfolio offers a comprehensive enterprise-grade solution that eliminates data silos, automates change tracking and synchronization, and integrates smoothly with AI workflows. It includes the NetApp AFX storage system and the NetApp AI Data Engine (AIDE). The AFX system provides scalable, disaggregated storage with a native parallel architecture for high performance. AIDE streamlines and secures the AI data pipeline through features such as data discovery, curation, guardrails, and vectorization for GenAI.

Built on NetApp ONTAP technology, the AFX AI portfolio is a component of the NetApp data platform. It provides a unified, enterprise-grade foundation for intelligent data infrastructure, ensuring consistent operations, seamless cloud integration, and embedded intelligence across various environments.

AFX and AIDE together deliver high-performance storage and intelligent data services, ensuring enterprise data is instantly AI-ready regardless of the deployment environment. This integration builds a robust foundation for subsequent business processes.

AI Evolution

As enterprise AI evolves from experimentation to measurable business results, organizations face the challenge of scaling due to fragmented data. This issue directly affects the operationalization of AI, prompting the development of advanced, innovative solutions.

Many organizations struggle with data silos, which hinder the efficiency and resilience of AI projects. Teams spend up to 80% of project time preparing data for AI, which increases costs and delays. Gartner predicts that inadequate AI-ready data will cause 60% of AI initiatives to be abandoned by 2026.

NetApp AFX AI Portfolio

The NetApp AFX AI portfolio provides enterprise-level performance, scalability, and resilience to support AI workloads from testing to deployment. It eliminates the need for specialized infrastructure silos, enabling seamless data access across on-premises and cloud environments without replication or re-platforming.

The core of the portfolio includes a global metadata catalog that lets users quickly identify, select, and publish data. It automates change detection and synchronization, ensuring that AI models have access to the latest enterprise data. In-place vectorization facilitates model training, fine-tuning, and retrieval-augmented generation (RAG) workflows without requiring additional tools.

The solution integrates security and governance throughout the data lifecycle by applying data tagging, masking, and policy enforcement. ONTAP supports resilient operations through immutable NetApp Snapshot copies and AI-enhanced Autonomous Ransomware Protection.

Native integration with leading AI platforms, including Domino, NVIDIA, and Informatica, enables compatibility with enterprise workflows. NetApp’s zero-copy caching and native cloud connectivity provide instant data visibility across any site, cloud, or model, so you can accelerate time to insight and enable agentic AI applications with confidence.

NetApp AFX Benefits

Infrastructure efficiency: Independently scale compute and capacity to match any AI workload, optimizing resources across hybrid multicloud environments.

High performance, low latency: Maintain consistent high speeds without the complexity of HPC, ensuring faster AI outcomes.

Smarter data operations: Enable instant indexing, semantic search, and in-place vectorization for rapid access and AI deployment.

Automated transformation: Reduce AI data storage needs by 5–10x with advanced data reduction technology, saving space and cost.

Built-in governance: Policy-driven guardrails automatically protect data privacy and compliance, ensuring trusted AI results.

Lower overhead: Use zero-copy caching, data reduction, and SnapMirror to cut costs and minimize data movement, simplifying management.

Certified storage for NVIDIA DGX SuperPOD: Proven, high-performance storage for the most demanding enterprise AI workloads.

Enterprise-ready: Get fast, scalable, secure, and managed storage—delivering all capabilities required for enterprise AI. Unified storage: Remove data silos, eliminate moves, and enable AI where your data resides.

NetApp AFX Specifications

AFX Cluster

Maximum storage controllers

128

Maximum storage enclosures

52

Maximum effective capacity

1+ EB

Maximum data compute nodes

10

Supported storage protocols

pNFS, NFS, SMB, S3, NFS/RDMA

AFX 1K Storage Controller

Controller form factor

2U

PCIe expansion slots

11 (9 + NVRAM)

X50131B—2p, 100G/200G/400G Ethernet

4

X50130B —

X50130B —

2p, 40G/100G Ethernet controller

100GbE ports (40Gbe autoranging)

5

X50133A—4p, 10G/25G Ethernet controller

10GBASE-T (1Gbe autoranging)

5

OS version

ONTAP 9.17.1 or later

Power consumption (median or typical)

Voltage – 200 (200 to 240)

Amps – 6.48 Typical

DX50 Data Compute Node

Form factor

1U

CPU

AMD Genoa 9554P

Sockets

1

Total cores

64

Memory

1TB

I/O

4-ports, 100GbE

GPU

1x NVIDIA L4

Internal storage

2xU.2 15TB

Typical power consumption

110 Voltage:

Typical – 4.31 AMPS

220 Voltage:

Typical – 9.10 AMPS

NX224 Storage Enclosure

Maximum drives

24

Drive form factor

2.5” small form factor

Form factor

2U

Drive type

supported

7.6 TB, 15.3 TB, 30.7 TB, 60TB

OS version

ONTAP 9.17.1 or later

Typical power consumption

TBD

Use Cases

RAG inferencing with real-time data. Automatically detect changes and synchronize enterprise data to support RAG workflows, so models operate on the most current and relevant information.

Confident model fine-tuning on enterprise data. Use secure, policy-driven data pipelines to fine-tune models with sensitive corporate data, for compliance, traceability, and performance.

Agentic AI orchestration. Enable autonomous agents to interact with enterprise data in real time, leveraging built-in guardrails, semantic search, and vectorized pipelines to support decision-making, task execution, and continuous learning.

AI factory enablement. Power scalable, repeatable AI workflows, from data ingestion to model deployment, using NetApp’s unified data foundation, integrated metadata services, and zero-copy caching to streamline operations across teams, clouds, and models.

Enterprise-Grade, High-Performance Disaggregated Storage

The AFX AI portfolio features NetApp AFX, a high-performance storage system designed for scalability, resilience, and security. Its disaggregated architecture allows independent scaling of compute and storage resources, optimizing AI infrastructure management.

Its natively parallel architecture leverages multi-threading and distributed computing techniques to efficiently handle large-scale data processing tasks. It supports standard file and object protocols such as NFS, SMB, S3, and REST APIs, eliminating the need for custom clients and simplifying deployment across diverse environments. Integrated cyber-resilience features include autonomous ransomware detection using heuristic and behavioral analysis algorithms, immutable backups stored on write-once read-many (WORM) storage media, and automated recovery workflows that coordinate with orchestrated restore procedures, providing robust protection for critical intellectual property.

Unified Data Foundation

ONTAP powers the AFX AI portfolio, giving enterprises a single view of data across on-premises systems and public clouds. NetApp uniquely offers native, fully integrated solutions on AWS, Azure, and Google Cloud, going beyond marketplace add-ons. NetApp ONTAP features:

• Centralized data visibility and management for hybrid and multicloud deployments

• First-party cloud integration ensures seamless data movement and consistent policies across all environments

• Unified governance, compliance, and operational efficiency by eliminating traditional data silos

• Automatic data tiering and replication between on-premises and cloud optimize performance and cost

A global metadata catalog and unified namespace ensure consistent visibility and management. Advanced data services like replication, deduplication, and compression improve data movement and storage efficiency. NetApp FlexCache enhances this by enabling zero-copy, real-time data sharing across different locations and clouds. This means AI workloads can access only the necessary data precisely when needed, quickly and securely, without copying entire datasets. Collectively, these features build a globally accessible, integrated data platform that supports scalable, secure, and efficient AI operations across any environment.

Accelerated Data Pipeline

NetApp AI Data Engine (AIDE) offers a comprehensive data service that makes AI simple, affordable, and secure, from preparing data to serving it to AI applications. Integrated into ONTAP, AIDE delivers a unified, real-time, structured overview of your entire NetApp data environment, enabling quick search and organization. It automates change tracking and data synchronization to reduce duplicates and incorporates safeguards that track data throughout the AI lifecycle.

By integrating automated vectorization for GenAI powered by NVIDIA NIMT, along with advanced compression, retrieval, semantic discovery, and secure policy-driven workflows, AIDE provides organizations with the efficiency, data clarity, and governance necessary to accelerate enterprise AI confidently.

The NetApp AFX AI portfolio, built on the NetApp data platform, delivers what modern AI requires: scalable performance, resilient and secure data management, seamless cloud integration, and an intelligent pipeline to operationalize AI workflows. It’s not just about enabling AI, it’s about making it work for the enterprise.

Engage with StorageReview

Newsletter | YouTube | Podcast iTunes/Spotify | Instagram | Twitter | TikTok | RSS Feed