Necessary Always Active
Necessary cookies are required to enable the basic features of this site, such as providing secure log-in or adjusting your consent preferences. These cookies do not store any personally identifiable data.
|
||||||
|
||||||
|
||||||
|
In Focus
Technology giant Cisco Systems has unveiled a new networking chip that is capable of connecting multiple AI data centers over long distances. Cloud computing divisions in Microsoft and Alibaba have already enlisted as buyers of the new Cisco P200 chip according to Yahoo Finance.
Cisco has also launched the Cisco 8223 router, a new routing device that is powered by the Cisco Silicon One P200 chip. Cisco touted the ability of its 8223 router to handle surging workloads while offering customers flexibility that allows AI firms to deploy models in agile infrastructure and evolve with the changing business needs.
The tech company said that the new router offers:
Data centers consume huge volumes of electricity. The demand for power has driven tech giants like Oracle and OpenAI to build data centers in Texas where they can access energy. Meta Platforms is setting up a data center in Louisiana as it searches for gigawatts.
Cisco’s inter-data center routing equipment becomes critical in connecting these facilities over vast distances. The company’s first quarter revenue estimates surpassed analyst projects after orders for AI infrastructure fueled growth.
The tech giant said the new 8223 router is designed to power next-generation AI and is the first 51.2-terabit fixed Ethernet router in the industry. Chipmakers like NVIDIA and AMD are connecting huge volumes of powerful computing chips in AI data centers.
“AI compute is outgrowing the capacity of even the largest data center, driving the need for reliable, secure connection of data centers hundreds of miles apart. With the Cisco 8223, powered by the new Cisco Silicon One P200, we’re delivering the massive bandwidth, scale and security needed for distributed data center architectures,” Cisco’s Executive VP for Common Hardware Group Martin Lund said.
The idea is to have the chips serve as a single brain in handling AI tasks. Cisco’s AI networking chip and router connect multiple data centers to create one huge computer.
“Now we’re saying, ‘the training job is so large, I need multiple data centers to connect together. And they can be 1,000 miles apart,” Lund told Reuters.
Cisco’s AI Networking System at a Glance:
The tech giant says its latest data center networking chip is highly compact. Built on Cisco’s Silicon One architecture, the new chip replaces one that took 92 separate chips into a single piece.
Cisco’s new router is designed to deliver deep-routing buffering, a technology that the company has worked on for decades. By leveraging buffering technology, the router syncs data across different data centers without losing sections of it.
“The increasing scale of the cloud and AI requires faster networks with more buffering to absorb bursts. We’re pleased to see the P200 providing innovation and more options in this space. Microsoft was an early adopter of Silicon One, and the common ASIC architecture has made it easier for us to expand from our initial use cases to multiple roles in DC, WAN, and AI/ML environments,” Corporate Vice President for Azure Networking at Microsoft Dave Maltz said.
With the new P200 chip, Cisco will be taking on networking firms like Broadcom which launched an Ultra Ethernet Switch in July 2025.