AI inference pushes SSD into the computing core! FADU Gen6 will be shipped in the second half of the year, reading 28.5GB/s, and Gen7 will be sampled in 2028

As generative AI moves from large-scale model training to ubiquitous inference applications, the next bottleneck in data centers is extending from GPU computing power to data access and movement. Tony Kim, Chief Commercial Officer of FADU (KOSDAQ: 440110), a Korean enterprise-level SSD controller manufacturer, said on the 20th that with the rapid increase in the amount of long text, AI Agent (AI Agent) and KV Cache (Key Value Cache) data, SSD is no longer just a back-end device that passively saves data, but has begun to become an AI computing...