NetApp launches high-speed data storage for AI factories

NetApp has launched a storage solution designed specifically for AI infrastructure providers operating large GPU environments. NetApp Novus has been designed to support a zettabyte-scale file system and an aggregate throughput of up to 100 TBps.

NetApp Chief Product Officer Syam Nair said storage architectures are increasingly becoming a bottleneck for AI factories being developed by neoclouds and GPU-as-a-Service providers, because the factories are operating at a scale the architectures were never designed to support. This can reduce GPU utilisation and increase infrastructure costs. GPU utilisation can drop below 30% using traditional storage architectures.

“AI factories struggle and GPU economics collapse when data can’t keep up. HPC-era storage alone cannot serve AI Factory scale — the industry needs a true architecture built for it from the ground up. AI factories have long been searching for infrastructure that can scale at AI speed,” he said. “Leveraging the power of the NetApp Platform, Novus delivers 100 TB/s of unlimited scale-out performance, leapfrogging the market to serve as the new high-performance, AI-scale storage architecture that unlocks the performance of millions of GPUs and delivers the return on investment needed to sustain the next wave of AI innovation.”

The initial release of the platform combines metadata software the NetApp Novus Data Detector running on Supermicro infrastructure with data services delivered through AFF A90 systems. This ensures metadata services are separated from the data path to reduce storage-related bottlenecks.

The architecture is designed to meet the demands of multi-tenant, cloud-scale AI factories and provide an eventual path to software-defined deployments in the same namespace, operating model and data foundation. Operators can scale only the resources needed for each workload to reduce power, space and infrastructure overhead.

Image credit: iStock.com/FlashMovie