DDN, Google Cloud Boost AI Storage to 10TB/s
As enterprises shift from AI experimentation to production deployment, storage infrastructure has become the critical bottleneck for GPU and TPU utilization. DDN and Google Cloud's Managed Lustre platform now delivers 10 terabytes per second throughput, a 10x increase, while achieving 95% or more TPU utilization for joint customers.
Real-world results include a 6x GPU saturation improvement for Harmonic and a 40% reduction in time-to-first-token for inference workloads. Customers including Salesforce and Sony Honda Mobility are already running production workloads on the platform, which both companies say is essential infrastructure for the agentic AI era.
