Open networking solutions vendor Edgecore Networks has thrown its hat into burgeoning 100-terabit data center switch ring.

Powered by Broadcom’s Tomahawk 6 line of chips, Edgecore’s AIS1600-64O and AIS800-128O lines can handle 102.4 Tb/s to meet the massive bandwidth and low latency needs of hyperscale AI clusters and cloud fabrics.

“We are delivering a platform that is not only faster in port speed but significantly more intelligent in traffic engineering and energy efficient for the AI-driven data center of 2026 and onwards,” PoWen Tsai, head of AI/cloud and the solution engineering business division at Edgecore Networks, explained.

The 100T+ switch market is becoming increasingly crowded, with vendors increasingly working toward solutions to tantalize hyperscalers eyeing the shift to 1.6 Tb/s.

Marvell Technology recently debuted the 102T Teralynx line, while Bill Gartner, SVP and GM for optical systems at Cisco’s Optics Group, recently told SDxCentral that customers are increasingly moving to 800 Gb/s (800G) and “most will move to 1.6T when that's available.”

Edgecore looks to be getting in on the action with a line of switches that meet increasingly connectivity intensity demands while supporting a broad range of open networking standards.

The AIS1600-64O is built for operators transitioning to 1.6T connectivity, with support for breakout configurations ranging from 100G to 1.6T. Built to a 3RU configuration, it packs some 64 OSFP1600 ports with the vendor touting it as suitable for supporting the inevitable launch of 800G network interface cards.

Edgecore’s AIS800-128O comes in a 4RU design and encompasses support for up to 512 logical ports on a single chip. The vendor claims it’s the highest density model available for massive fabrics connectivity, with it capable of supporting the largest two-tier scale of 400G NIC connectivity.

Beyond raw speed, both feature capabilities to reduce job completion time (JCT) through hardware-based link failover and intelligent load balancing (DLB/GLB).

“The transition to 1.6T Ethernet is a critical milestone in meeting the relentless bandwidth demands of next-generation AI clusters,” Hasan Siraj, VP of product management for Broadcom’s Core Switching Group, explained. “By utilizing our latest Tomahawk 6 silicon, Edgecore is delivering a high-performance, power-efficient platform to the open networking ecosystem that empowers customers to scale their infrastructure with confidence.”

“These solutions reinforce our steadfast commitment to open networking technology and solutions, ensuring our partners and customers continue to realize the benefits of hardware-software disaggregation through greater innovation with flexibility and rich ecosystem choices,” Tsai added.

Edgecore joining the race for space to supply solutions to appease hyperscaler appetite for AI back-end buildout comes as Dell’Oro Group data projects the market to surpass $100 billion by 2030.

The analyst firm suggests that the back-end boom represents a lucrative opportunity for switch suppliers, particularly with massive scale-out (cluster-to-cluster interconnectivity) and scale-across (distributed data centers) projects like Amazon Web Services; (AWS’) Rainer or Microsoft’s Fairwater driving demand for high-performance infrastructure solutions.