Alibaba, Apple, and Synopsys have joined the Ultra Accelerator Link (UALink) Consortium.
In a statement, UALink said the new board members will “share their industry knowledge for development and adoption of UALink – performance.”
Billing itself as an alternative to Nvidia’s NVLink offering, UALink was founded in May 2024 to define and establish an open industry standard that will enable AI accelerators to communicate more effectively and efficiently.
Current consortium members include AMD, Intel, Meta, Hewlett-Packard Enterprise, Amazon Web Services (AWS), Astera Labs, Cisco, Google, and Microsoft, with photonic computing company Lightmatter also joining the group at the start of January 2025.
“We are pleased to welcome Alibaba, Apple, and Synopsys to the UALink Consortium Board of Directors,” said Kurtis Bowman, UALink Consortium board chair. “Since our incorporation, the Consortium has grown to more than 65 total members, spanning cloud, silicon and IP providers, software companies, system OEMs, and others.”
In December 2024, Synopsys unveiled a UALink IP solution that will offer up to 200 Gbps throughput per lane, linking up to 1,024 accelerators. The UALink 1.0 specification will be publicly available in Q1 2025 with Synopsys’ solution scheduled to be available in the second half of 2025.
Speaking to DCD when the IP solution was announced, Priyank Shukla, principal product manager for interface IP at Synopsys said that there was a need for an open standard that connects accelerators at scale as by 2030, it's estimated that 70 percent of all compute will be for AI accelerator training.
“We really need efficient interconnects for that infrastructure and both of these protocols together provide very efficient, both power efficient and bandwidth efficient, and latency-sensitive interconnect technology for accelerator to accelerator networks,” he said.
