_2_(1).svg?width=5760&format=avif&disable=upscale)
NETAPP AFX
NetApp® AFX is high-performance, disaggregated storage built for enterprise AI. Extreme throughput
and exascale capacity meet the enterprise-proven reliability of ONTAP®: one secure data foundation
for every stage of your AI pipeline.
AI innovators across industries trust NetApp
ASML Holding
LUMEN
San Francisco
BOSCH
NVIDIA
Kyndryl
Anaplan
Cadence
SK Telecom
Domino Data Lab
Ducati
World Wide Technology
WHAT IS NETAPP AFX?
NetApp AFX separates the compute layer (storage controllers) from the capacity layer (NVMe enclosures), letting you scale performance and capacity independently. Built on NetApp ONTAP, it delivers extreme throughput, exascale capacity, and enterprise-grade security without the complexity of traditional parallel file systems or proprietary clients.
Why AFX for enterprise AI?
AFX is engineered to supercharge your entire AI data pipeline. Eliminate the complexity and compromise of traditional AI storage.
You need the power of HPC-class storage with the agility to adapt to rapidly evolving AI workloads without maintenance windows or performance-draining rebalancing. Our early-access customers told us they can grow their clusters as needed while AFX auto-balances resources without affecting performance. These same customers put our storage through its paces with resilience testing while under heavy I/O workloads, and we passed with flying colors.
As enterprises embrace agentic AI, one truth is becoming clear: Security can’t be an afterthought. NetApp ONTAP, the world’s most secure storage platform, gives you a strong foundation for enterprise AI. Our integrated NetApp Autonomous Ransomware Protection with AI, designed for better than 99% accuracy, keeps your datasets and models safe from even the most sophisticated threats.
Hybrid cloud operations are the norm for the modern enterprise, and you need AI everywhere your data lives. Thanks to native ONTAP integrations and first-party services on AWS, Azure, and Google Cloud, you can streamline your AI pipelines with unified data access across on-premises and cloud environments. This makes it simple to manage, protect, and mobilize data using ultra-efficient replication and intelligent caching.
How NetApp AFX architecture provides differentiated value
The disaggregated NetApp AFX architecture enables capacity and compute resources to scale independently and node performance to scale linearly.
With hardware resources decoupled in the NetApp AFX architecture, nodes no longer require associated disks to be added alongside one another. When a cluster is running low on performance-related resources such as RAM, CPU, or network throughput, only storage nodes need to be added to the cluster and can leverage the existing Storage Availability Zone. Conversely, if the requirement is capacity, only shelves would need to be added to the mix.
NetApp AFX replaces per-node physical aggregates with a single Storage Availability Zone available to the entire cluster.
In NetApp AFX, nodes and shelves are all connected to the same backend switch, which makes it possible for ONTAP to extend the overall visibility domain for disks to the entire stack. As a result, no nodes own any specific disks. Rather, all disks participate in a single capacity pool called a Storage Availability Zone, which provides simpler capacity management and increased performance potential. (More available disks means more available performance.)
The enterprise-grade data platform for AI
NetApp AFX supports the same networking stack, security features, data protection technologies, nondisruptive operations, and volume types as NetApp ONTAP, with some adjustments unique to the disaggregated architecture.
Stat: 4 TB/s
THROUGHPUT FOR AI-
INTENSIVE WORKLOADS,
WITH INDEPENDENT AND
LINEAR SCALING OF
PERFORMANCE AND
CAPACITY*
Note: *Throughput for AI-intensive workloads, with independent and linear scaling of performance and capacity
Stat: 1+ EB
SCALING CAPACITY
FOR MASSIVE DATA
SETS**
Note: **Up to 1+ EB in a single cluster with FabricPool tiering
Stat: 99%
ACCURACY WITH REAL-TIME
RANSOMWARE DETECTION***
Note: ***SE Labs testing results in 99% ransomware detection accuracy and 100% precision accuracy rating for file workloads
NetApp AFX was built with performance and scale in mind – specifically geared towards workloads that require high read and write throughput and can provide simple, linear scale.
AFX is ideal for training LLMs and AI at up to 4 TB/s and scales to an exabyte, so GPUs never wait on storage. AFX is certified for NVIDIA DGX SuperPOD and production ready from day one.
Fast compute isn't enough; inference needs fast memory. AFX puts a low-latency KV cache tier next to the GPU and AI Data Engine (AIDE) indexes metadata in real time. The result: faster first token and lower cost per query.
AFX gives autonomous AI agents governed access to one
unified namespace across on-premises and hybrid cloud.
Guardrails, secure multi-tenancy, and QoS keep agents
scoped to the data you intend. No sprawl. No exposure.
NetApp AFX disaggregated storage and NetApp AI Data Engine (AIDE) provide a unified foundation
that powers your AI factory with scalable performance, built-in governance, and real-time data
intelligence. Together, they accelerate enterprise AI, from initial training to production deployment.
AFX technology delivers disaggregated storage with proven high availability, so organizations can scale AI infrastructure efficiently while maintaining enterprise-grade resilience, security, and data management.
Simplify and secure AI data pipelines with
NetApp® AI Data Engine™. Get unrivaled data
clarity, efficiency, and governance from data
preparation to GenAI app deployment.
Action: Meet NetApp AIDE /platform/ai-data-engine
Customer stories
Read customer success stories that demonstrate the value of an intelligent data infrastructure built for AI.
Attributed to: Jagdish Joshi, Principal of Technology Infrastructure, Mahindra
Source: Mahindra
Attributed to: Mike Valora, Technical Solutions Architect, World Wide Technology
Source: World Wide Technology
Attributed to: Kevin Cochrane, CMO, Vultr
Source: VULTR
TECHNICAL RESOURCES
NetApp AFX systems are ideal for data-intensive workloads like AI that demand maximum throughput and
performance. Pair with the NetApp AI Data Engine to accelerate your AI adoption.
https://www.netapp.com/pdf.html?item=/media/142853-ds-3466-netapp-afx-datasheet.pdf
https://www.netapp.com/pdf.html?item=/media/159595-nva-1184-superpod-gb300-afx-design-guide.pdf
https://www.netapp.com/blog/platform-ai-data-infrastructure/
_1.svg){kind=link}
