NetApp Novus has AI factory bottlenecks in the crosshairs
NetApp's zettabyte-scale Novus storage system targets AI factory I/O bottlenecks—the infrastructure chokepoint nobody's talking about while everyone focuses on compute.
Why it matters
AI factories are compute-bound in theory but often I/O-bound in practice. Novus is a production storage architecture designed to handle AI workload data throughput at scale, with concrete implications for on-prem and cloud AI deployment economics.
The key facts
5 to knowNetApp Novus announced as zettabyte-scale file system
Targets AI factory data pipeline bottlenecks, not compute
Traditional storage architectures cited as inadequate for AI workload throughput
Positions storage as a buildout constraint in AI infrastructure
No pricing, availability date, or measured throughput vs. competitors disclosed in article
Go to the source
ITProitpro.com
Publisher excerpt: The new zettabyte-scale system aims to overhaul traditional storage architectures