Building on its award-winning large-radix (256-lane) CXL and PCIe “Apollo” Switch, XConn is poised to drive UALink switch innovation for emerging AI ecosystem
SANTA CLARA, Calif., Nov. 12, 2024 /PRNewswire/ — XConn Technologies (XConn), the innovation leader in next-generation interconnect technology for the future of high-performance computing and AI applications, today announced it has joined the Ultra Accelerator Link™ (UALink™) Consortium as a Contributor member. The newly incorporated UALink Consortium aims to establish an industry-standard, high-speed interconnect to enable scale-up communications between AI accelerators and switches, delivering low latency and high bandwidth for AI workloads in data center environments.
XConn’s participation in the UALink Consortium follows its successful introduction of the “Apollo” switch, a pioneering 256-lane CXL and PCIe switch fabric architecture, which has set new standards for scalability and flexibility in data center interconnect solutions. Leveraging this large-radix architecture, XConn is uniquely positioned to contribute to the development of UALink’s high-performance interconnect, which will play a critical role in supporting the expanding needs of AI and HPC workloads.
“We are thrilled to contribute to the development of an open standard that directly addresses the high-performance needs of AI infrastructures,” said Gerry Fan, CEO of XConn Technologies. “Our award-winning large-radix Apollo switch architecture positions XConn as a leading force in enabling the scale and efficiency required for the UALink ecosystem. We look forward to working with industry leaders to build a comprehensive ecosystem that empowers future AI deployments and fosters scalable innovation across data centers.”
The UALink Consortium was established by a coalition of leading technology companies, including AMD, Amazon Web Services (AWS), Astera Labs, Cisco, Google, Hewlett Packard Enterprise (HPE), Intel, Meta, and Microsoft. The Consortium’s members are engaged to help shape and refine the UALink 1.0 specification, which will be available for review by Consortium members this year and for general release in Q1 2025. The UALink 1.0 specification is designed to support high-speed data transfer of up to 200Gbps per lane and can connect up to 1024 accelerators within an AI pod. This enables data centers to achieve the efficiency and scalability required for large AI models and compute-intensive applications.
“We are pleased to welcome XConn Technologies as a Contributor member,” said Willie Nelson, President of the UALink Consortium. “With their expertise in interconnect technology, XConn brings a valuable perspective to our mission of developing an open, high-performance standard that can meet the demands of emerging AI and high-performance computing applications.”
More information on the UALink Consortium, which announced its incorporation last month, is available at www.ualinkconsortium.org.
About XConn Technologies
XConn Technologies Holdings, Inc. (XConn) is the innovation leader in next-generation interconnect technology for high-performance computing and AI applications. The company is the industry’s first to deliver a hybrid switch supporting both CXL 2.0 and PCIe Gen5 on a single chip. Privately funded, XConn is setting the benchmark for data center interconnect with scalability, flexibility and performance. For more information visit: xconn-tech.com.
About Ultra Accelerator Link Consortium
Ultra Accelerator Link (UALink) is a high-speed, scale-up accelerator interconnect technology that advances next-generation AI cluster performance. AMD, Astera Labs, AWS, Cisco, Google, HPE, Intel, Meta and Microsoft formed an open industry standard body, Ultra Accelerator Link Consortium, to develop technical specifications that facilitate breakthrough performance for emerging usage models while supporting an open ecosystem for data center accelerators.
Media Contacts:
Steve Sturgeon
Avista PR
ssturgeon@avistapr.com
858.472.5669
SOURCE XConn Technologies