In focus
Supermicro, IBM and NVIDIA Unveil KV Cache Storage Solution for AI Inference
Supermicro has launched a KV cache storage solution built with IBM and NVIDIA, combining Petascale storage servers, IBM Storage Scale software, NVIDIA Dynamo and Spectrum-X networking. Joint testing showed up to 56x faster time-to-first-token at 130,000-token prompts.
Read the story