Cisco introduced the Silicon One G300, a 102.4 Tbps switching silicon targeting massive artificial intelligence cluster buildouts in modern data centers. The new chip is positioned as the core of next generation networking infrastructure optimized for artificial intelligence workloads, aiming to turn the network into an artificial intelligence innovation platform rather than just a transport layer.
The Cisco Silicon One G300 will power new Cisco N9000 and Cisco 8000 systems that are intended to push the frontier of artificial intelligence networking in the data center. These systems incorporate liquid cooling and support high density optics to establish new efficiency benchmarks and help customers maximize returns on expensive GPU investments by removing bandwidth and cooling bottlenecks at the network layer.
Alongside the new silicon and hardware platforms, Cisco enhanced its Nexus One offering to simplify how enterprises operate artificial intelligence networks on premises or in public cloud environments. The updates focus on reducing operational complexity that can prevent organizations from scaling artificial intelligence data centers, with Cisco emphasizing full stack innovation from silicon to systems and software to support customers ranging from hyperscalers to traditional enterprises as they shift to artificial intelligence powered workloads.
