DDN, Google Cloud Claim Lustre Kv Cache Trick Boosts AI Inference Throughput 75%
sdxcentral, Friday, April 24th, 2026
DDN added new capabilities to the Lustre platform it manages with Google Cloud, including means to share key-value (KV) cache to boost AI inference workloads.
Unveiled at Google's annual Next event, the pair showcased using Managed Lustre as a shared cache layer across inference clusters rather than keeping KV-cache in each server's local memory. As a result, DDN and its hyperscaler helper claim total inference throughput was improved by as much as 75%.
The parallel file system made its debut in 2025, and lets users deploy and scale Lustre environments for high-performance computing (HPC) and AI workloads without the need to manage the underlying servers.