Arista Announces Rack-Scale Architectures for Scale-Up and Scale-Out AI Fabrics
Business Wire•07/10/2026•09:00 ET
9
Key Highlights
- ➤Etherlink SU-144 supports 144 accelerators per rack, scaling to 1,024 XPUs cross-rack
- ➤Next-generation optics lift rack capacity to 6.5 Pb/s from 1.6 Pb/s
- ➤Arista introduces open Ethernet rack-scale architectures for AI scale-up and scale-out fabrics
- ➤New rack designs support 100–400kW liquid cooling for extreme AI density
- ➤Arista’s next-generation optics nearly halve overall data center footprint
Expert Statements
Tyson Lamoreaux, Senior Vice President, Cloud and AI Networking at Arista Networks
“The network in the AI era is no longer just an independent pipe between servers. It’s now a tightly integrated rack-scale backplane that is essential to maximizing compute”
Tyson Lamoreaux, Senior Vice President, Cloud and AI Networking at Arista Networks
“Arista is delivering an open, Ethernet-based framework for both scale-up and scale-out fabrics. By combining liquid cooling solutions with the choice of Arista EOS® or open NOS, we empower cloud titans and enterprise innovators to scale heterogeneous compute and optimize their token economics without proprietary lock-in.”
Arista Announces Rack-Scale Architectures for Scale-Up and Scale-Out AI Fabrics
Engineered with AMD, Arm, Broadcom, d-Matrix, Meta, Microsoft, and Qualcomm to deliver extreme compute and networking density for AI
Get started
Create a free account to read the full story.