NetApp Removes Storage Bottleneck for AI Factories
100 TB/sec. NetApp's Novus storage just rewrote the I/O bottleneck for GPU clusters running agentic workloads.
Why it matters
NetApp Novus is purpose-built storage architecture for AI factories — designed to maximize GPU utilization by eliminating I/O wait time at zettabyte scale. This matters to infrastructure operators and cloud providers building production agent systems where compute sits idle waiting for data.
The key facts
6 to knowNetApp Novus announced September 30, 2026
Designed for up to 100 TB per second aggregate throughput
Architected for zettabyte-scale storage operations
Purpose-built for large GPU environments and AI infrastructure providers
Focused on maximizing GPU utilization by reducing storage bottlenecks
GA status not explicitly stated in excerpt; full product details truncated in source
Go to the source
EnterpriseAIhpcwire.com
Publisher excerpt: New architecture delivers fastest storage on the planet, designed to maximize GPU utilization and architected for up to 100 TB per second aggregate throughput SAN JOSE, Calif., Sept. 30, 2026 — NetApp has announced NetApp Novus, the fastest storage on the planet designed for AI infrastructure…