NEW CASTLE, DE — FS launched an 800G Ethernet networking platform designed for NVIDIA B300 GPU clusters, positioning the company to compete in the rapidly expanding market for AI data center infrastructure as demand for high-bandwidth computing accelerates. The company said the new platform uses a 51.2-terabit RoCEv2 Ethernet architecture intended to support large-scale artificial intelligence training workloads that require low-latency and loss…
This story is only covered by news sources that have yet to be evaluated by the independent media monitoring agencies we use to assess the quality and reliability of news outlets on our platform. Learn more here.