Keysight Sets New Benchmark for Silicon Validation of AI Ethernet Switching at 51.2T

Validates the performance of next-generation data center switches used in advanced AI and accelerated computing environments

Keysight Technologies, Inc. has achieved a new milestone in validating the accelerated infrastructure necessary to power the next-generation AI-driven cloud data center with Keysight’s AresONE-M 800GE test platform and the Marvell Teralynx 10 Ethernet switch chip.

“To definitively measure and validate the low-latency capabilities of Teralynx 10, Marvell turned to Keysight and its industry-leading AresONE-M 800GE test equipment. Their thorough evaluation and collaboration with our team made the process smooth and ensured the integrity of the strong results we achieved.”

AI networks require high performance, unprecedented scale, and low-latency switching fabric to ensure seamless handling of massive volumes of data and real-time decision-making that advanced AI workloads demand. Keysight collaborated with Marvell Technology, Inc. to address the unique testing needs of network equipment manufacturers, silicon chipset vendors, and data center operators as they equip, build, and manage high-speed 800GE and 1.6T networks. Implementing an effective validation strategy helps mitigate the risk of costly product re-designs, recalls, or negatively impacting customer loyalty by assuring solutions will perform as designed and promised in real-world environments.

Recommended AI News: Rivet Revenue Diagnostics Diagnoses Financial Wellness for Healthcare

The milestone testbed was comprised of Keysight’s AresONE-M 800GE, the industry’s most comprehensive multi-speed Ethernet performance test platform and traffic generator, and the Marvell Teralynx 10 Ethernet programmable switch device, which offers industry-leading low latency, supporting up to 51.2 Tbps throughput with comprehensive analytics.

Highlights from the validation include:

• High throughput and high speed – The Keysight AresONE-M 800GE generated 51.2 Tbps of traffic, sending it successfully through the Teralynx 10 switch to test its limits. The test validated an 800GE interface speed based on 112G SerDes, which facilitates faster data transfer between devices and extended reach in data center interconnects or telecommunications networks that run data-intensive AI applications.
• Unprecedented scalability – Eight AresONE-M 8-port chassis were chained together, providing an industry first 64 x 800GE link configuration to achieve a high scale test bed running at 800GE line rate.
• Low latency – Low latency is critical for achieving the shortest job completion time for AI training and other highly distributed applications. These AI workloads depend on the switching fabric to provide the lowest possible latency — and to do so predictably.
• Performance analysis – Beyond the need for high bandwidth and low latency, measuring the performance of all 64 x 800GE links was an important aspect of the testbed, providing deeper, actionable analytics around loss, latency, and jitter at line rate. These valuable insights will help accelerate innovation for data centers deploying future AI applications.

Recommended AI News: CDNetworks Upgrades WAAP Solution with Its Latest AI-Powered Cloud Security 2.0 Platform

Rishi Chugh, Vice President of Product Marketing, Network Switching, Marvell, said: “To definitively measure and validate the low-latency capabilities of Teralynx 10, Marvell turned to Keysight and its industry-leading AresONE-M 800GE test equipment. Their thorough evaluation and collaboration with our team made the process smooth and ensured the integrity of the strong results we achieved.”

Ram Periakaruppan, Vice President and General Manager, Network Test & Security Solutions, Keysight, said: “The switching fabric is a vital component of the back-end network architecture used for AI training. AI training workloads trigger a huge increase in traffic volume and compared to front-end networks, are persistently pushing this higher traffic volume around the clock. At these sustained rates, benchmarking low latency becomes very critical for ensuring that the AI training algorithms achieve the most efficient job completion time.”

Recommended AI News: Secureworks Threat Score Ushers In A New Age Of Cybersecurity AI

[To share your insights with us as part of editorial or sponsored content, please write to sghosh@martechseries.com]