Intel positions InfiniBand networking as critical alternative for AI datacenter high-speed interconnect

Cornelis Networks, an Intel spin-off launched in 2020, announced its new 400-Gbps CN5000 family of scale-out networking solutions on Tuesday. CEO Lisa Spelman told CRN that the products offer the "best performance with devastatingly good price-performance" compared to offerings based on InfiniBand or Ethernet, including those made by Nvidia. The company, which spun out roughly a year after Intel halted development of its Omni-Path architecture for data centers running high-performance computing and AI workloads, positions the CN5000 family as a solution to help companies save money and increase overall performance as AI models grow larger and more complex, demanding increasingly faster and more expensive data center infrastructure.

Spelman, a former Intel executive who became CEO of the channel-driven company last year, said Omni-Path is what enables the CN5000 networking products to deliver two times higher message rates, 35 percent lower latency and up to 30 percent faster performance for high-performance computing workloads like computational fluid dynamics compared to Nvidia's 400-Gbps InfiniBand NDR solution. Equipped to support data center deployments of up to 500,000 nodes, the CN5000 family—which includes SuperNICs, switches, cabling and management software—can also deliver six times faster collective communication for AI applications than RDMA over Converged Ethernet (ROCE) solutions. These improvements, made possible by features like credit-based flow control and dynamic fine-grained adaptive routing, translate to improved GPU and processor utilization. "I really don't think as many people as you would think truly understand how underutilized some of that compute is," Spelman said.

A particular achievement for the company is that the ASICs underlying the CN5000 switch and SuperNIC products are the first versions of the chips that have been manufactured. "I've been in the industry 25 years, and this is my first time being part of a team delivering first-pass, production-quality silicon success," Spelman said. "And it just shows the commitment to quality, to engineering excellence, to use of modern tools, including AI [and] emulation—all of that for our design phase, our modeling phase [and] our validation phase." The CN5000, which will start shipping to customers this month and become broadly available in the third quarter, "seamlessly interoperates" with CPUs, GPUs and accelerator chips from AMD, Intel, Nvidia and other companies.

While the CN5000 products are aimed at on-premises deployments for enterprise, government and academic customers running AI and HPC workloads, Cornelis sees substantially bigger market opportunities with the next two generations. With next year's 800-Gbps CN6000 products, Cornelis plans to enable Ethernet compatibility with ROCE support, developed with a "cloud-definitional customer," that will allow Ethernet-based networks to access some of Omni-Path's features. "It was the innovation around creating the adaptation layer that we've built in that allows you to get access to those [Omni-Path] features, because otherwise we're just another standard Ethernet vendor banging our head against a big wall," Spelman said. Then in 2027, the company plans to release its 1.6-Tbps CN7000 products, which will integrate standards from the Ultra Ethernet Consortium into Omni-Path and feature support for 2 million nodes and in-network computing.