NetApp has introduced two products geared toward accelerating artificial intelligence (AI) and data analytics in large-scale data center environments: the NetApp AFX all-flash storage system and the NetApp AI Data Engine. These new offerings aim to modernize enterprise data platforms for high-performance, mission-critical AI tasks. Both products are available for direct purchase or through subscription via NetApp Keystone Storage-as-a-Service.
According to NetApp, the AFX system is an enterprise-grade, disaggregated flash storage architecture based on NetApp ONTAP. It is designed for demanding AI workloads and supports linear performance scaling up to 128 nodes. NetApp reports AFX can deliver terabytes per second of bandwidth and exabyte-scale capacity, with the ability to independently scale performance and capacity. AFX is certified for NVIDIA DGX SuperPOD supercomputing and includes secure multi-tenancy, robust data management, and seamless integration with both on-premises and cloud infrastructure. Optional DX50 data compute nodes add a real-time metadata catalog and leverage NVIDIA accelerated computing.
The NetApp AI Data Engine operates as a comprehensive service for the AI data pipeline, handling ingestion, preparation, and deployment of AI workloads. It provides a global view of all NetApp-managed data, automates change detection, ensures current datasets, and applies built-in security guardrails. The AI Data Engine runs on AFX clusters and DX50 data compute nodes, integrating the NVIDIA AI Data Platform reference design, which includes NVIDIA accelerated computing and NVIDIA AI Enterprise software with NIM microservices for vectorization and retrieval tasks. NetApp notes that planned ecosystem support will add integration with NVIDIA RTX PRO Servers using RTX PRO 6000 Blackwell GPUs.
NetApp has also announced an Object REST API, now in public preview, for Azure NetApp Files. This API allows direct connection of NFS and SMB datasets to Microsoft Azure AI and data services—such as Azure OpenAI, Microsoft Fabric, Azure Synapse, and Azure Databricks—without needing to move or copy file data to a separate object store. In addition, new FlexCache capabilities enable global namespace unification between on-premises and Azure environments. Enterprises can migrate data and snapshots between environments using SnapMirror to support hybrid cloud use cases like backup and disaster recovery.
Syam Nair, Chief Product Officer at NetApp, stated, “With the new NetApp AFX systems, customers now have a trusted, proven choice in on-premises enterprise storage built on a comprehensive data platform to rapidly propel AI innovation forward,” and described how the technology allows organizations to connect their data estate across hybrid multicloud environments, collapsing multiple data management steps into NetApp’s integrated AI Data Engine. Justin Boitano, Vice President, Enterprise AI Products at NVIDIA, commented, “NetApp’s data platform has transformed into an AI-native storage platform by integrating NVIDIA accelerated computing and software, including leading AI models. With this new platform, organizations can index and search vast amounts of unstructured data across their enterprise to drive innovation and deliver real business impact.”
Source: NetApp
