Cisco powers the scale-across era\u00a0<\/p>\n
AI workloads are growing faster than the data centers that host them.\u00a0 AI training workload demands are effectively limitless, constrained only by power availability. To keep up, data centers are migrating to remote locations, wherever power is available. This migration requires unprecedented bandwidth to interconnect massive AI clusters over large distances. These huge AI training runs can last months and are incredibly costly, and any network downtime risks leaving a billion-dollar data center idle.\u00a0\u00a0\u00a0<\/p>\n
To meet the demands of AI workloads, data center networks have been scaling along two dimensions \u2013 scale-up, connecting GPUs within a rack, and scale-out, connecting racks within a data center. Now, a third dimension is essential to unlock new AI intelligence: \u201cscale-across\u201d extends AI workloads across multiple data centers while making them operate as one coherent, unified cluster. To achieve this, networks must be ultra-reliable, secure, high bandwidth, and scalable \u2013 effectively merging routing and switching requirements.\u00a0<\/p>\n
That\u2019s why we built the Cisco 8223<\/a>, powered by Cisco Silicon One P200<\/a>. It\u2019s the industry\u2019s first 51.2-terabit fixed Ethernet router, featuring the highest bandwidth 51.2-terabit deep-buffer routing silicon. The Cisco 8223 is purpose-built for scale-across networking demands. With industry-leading power efficiency, scalability, and security features, it\u2019s engineered to handle what\u2019s next.\u00a0\u00a0<\/p>\n