Skip to content
Featured on Startup Fame

NetApp launches Novus storage architecture for AI metadata, acquires PEAK:AIO

NetApp has launched Novus, a storage architecture built to remove metadata bottlenecks in AI workloads, and announced its planned acquisition of PEAK:AIO Ltd. to enhance parallel file system capabilities. The moves aim to improve scalability and performance for enterprise AI applications without requiring customers to rebuild existing infrastructure.

Editor, Lazyfounder

Published Updated 3 min read
NetApp launches Novus storage architecture for AI metadata, acquires PEAK:AIO
Image: SiliconANGLE via source

Novus architecture targets AI metadata bottlenecks

NetApp has unveiled Novus, a storage architecture designed to eliminate metadata bottlenecks in AI data infrastructure. The system delivers over 100TB/s of aggregate throughput, enabling enterprises to scale GPU clusters without rebuilding their storage layer.

The architecture separates metadata management from stored data using a component called Data Director. This allows independent scaling of metadata, which NetApp says is the key to supporting zettabyte-scale AI workloads. Applications gain a unified view of files across multiple ONTAP storage clusters, reducing the need for direct cluster-by-cluster access.

PEAK:AIO acquisition aims to simplify integration

Alongside Novus, NetApp announced plans to acquire UK-based PEAK:AIO Ltd. The deal will bring parallel file system technology built on the pNFS protocol, which is already supported by any Linux kernel released after 2018. The goal is to integrate PEAK:AIO’s capabilities without requiring customers to replace existing infrastructure.

PEAK:AIO’s technology is expected to complement Novus by enabling high-performance file access across distributed environments. NetApp’s Chief Product Officer Syam Nair emphasized that the combined solution will allow enterprises to extend their current storage systems rather than undergo costly migrations.

AI Data Engine accelerates data preparation

NetApp introduced its AI Data Engine, a tool designed to prepare enterprise data for AI applications. The engine automates discovery, classification, and vectorization of data, reducing what the company says typically takes six to nine months of manual engineering work to a fraction of that time.

The company also highlighted its unified approach to storage, supporting file, block, and object protocols. This flexibility is intended to help enterprises use data where it resides—whether on-premises, at the edge, or in the cloud—without creating new silos.

Expanding cloud partnerships and data governance

NetApp’s broader strategy, announced at NetApp INSIGHT, focuses on making legacy data AI-ready without infrastructure overhauls. The company plans to launch a fully managed storage service on Oracle Cloud Infrastructure within the next year, building on its existing ONTAP integration with Oracle. This aligns with its earlier co-engineered storage service for OCI, which aimed to simplify AI workload migration and modernization.

The company has also expanded its partnership with Commvault Systems for data protection and introduced Console, a control plane that provides fleet manageability, observability, and analytics. Console can be deployed locally for air-gapped environments, addressing compliance and security requirements.

Building on recent NetApp-Oracle collaboration

Last month, NetApp and Oracle launched a native ONTAP storage service in Oracle Cloud Infrastructure, a co-engineered solution designed to streamline AI workloads and infrastructure modernization. The new fully managed service announced this week extends that collaboration, aiming to simplify deployment for enterprises already using OCI.

Earlier this year, concerns over CoreWeave’s debt levels triggered a selloff in AI and cloud stocks, including Oracle. NetApp’s continued cloud expansion with Oracle signals confidence in hybrid and multi-cloud strategies, even as market sentiment fluctuates around infrastructure providers.

What this means

Lazyfounder analysis — our interpretation, not reported fact.

NetApp’s Novus architecture and the planned acquisition of PEAK:AIO mark a direct response to the metadata bottlenecks that have slowed AI training at scale. For founders and operators running GPU clusters, this move eliminates a major friction point: instead of rewriting pipelines to fit legacy storage, they can slot Novus into existing ONTAP environments and scale metadata independently. The separate Data Director layer means governance, classification and vectorisation now run asynchronously, cutting months of data prep down to weeks without forcing a rip-and-replace. With Oracle Cloud and Commvault already integrated, NetApp is offering a pre-validated stack that reduces the risk of multi-vendor misconfiguration—a tangible advantage over DIY parallel file systems or single-cloud alternatives.

Sources

  1. SiliconANGLE · 2026-10-05
    NetApp’s Novus tackles metadata bottlenecks in AI data infrastructure
  2. SiliconANGLE · 2026-10-06
    NetApp aims to make legacy data AI-ready without a rebuild

This story is an original summary drafted with AI by Lazyfounder from the reporting listed above and checked by automated validation. Facts are attributed to their original publishers; sections marked as analysis are Lazyfounder's. Where a source is in another language, facts were machine-translated and quotations are reported, not reproduced. Read the original coverage via the links, and see our AI policy and corrections policy.

About the author

Editor, Lazyfounder

Tarun Mottlia edits LazyFounders, covering Indian startups, funding rounds, AI and product launches. Every story on the site is AI-assisted and checked against its cited sources before publication.

More stories by Tarun Mottlia

Get the LazyFounder Brief

Startup, funding and AI news in a five-minute read. Join the early-access list.

Lazy Founder - Powered by Blogy.in

Contact us

Have a story tip, correction or partnership idea?

Write to us at tarun.kumar@blogy.in or talk to the founder directly. We read every message.

Contact us