Ultra Accelerator Link™ (UALink™) is an open industry standardized Interconnect for Accelerator-to-Accelerator communication.
The UALink Consortium is an open industry standard group formed to develop interconnect technical specifications that facilitate direct load, store, and atomic operations between AI Accelerators (i.e. GPUs). Their focus is to enable low latency/high bandwidth fabric for hundreds of accelerators in a pod and to enable simple load and store semantics with software coherency. The UALink 1.0 Specification will leverage the experience of the Promoter Members who have developed and deployed a broad range of accelerators and switches.​​
​
Consortium companies represent a wide range of industry expertise, including Cloud Service Providers, System OEMs, Accelerator developers, Switch developers, and IP providers. Work is currently underway to develop additional usage models for data center AI connectivity.