Dedicated solution addresses AI’s unique testing needs and complexity
Spirent Communications has announced the industry’s first high-density test solution capable of emulating realistic Artificial Intelligence (AI) workloads over Ethernet. As the networking world’s technology of choice, Ethernet is the backbone of the Cloud and the ability to emulate realistic AI traffic workloads and test their impact on AI data center networks and interconnects is key.
“AI is an important focus for Spirent as applications such as ChatGPT, Lensa, Copilot and others are changing the communications landscape, and hyperscale cloud providers shift investment from their traditional data center front-end focus to new back-end infrastructures needed to manage the explosion in AI applications and workloads,” said Eric Updyke, CEO, Spirent Communications. “These new environments are increasingly being built and operated separately from traditional data centers and are physically very different in order cope with the specific needs of AI.”
Communications networks worldwide rely on Ethernet technology, and testing how the Ethernet fabric of new AI environments performs has proved to be an enormous challenge. AI training brings unique requirements, with dramatically increased workload volumes, high sensitivity to latency, congestion, and job completion times (JCT).
“We are committed to helping customers bring new AI solutions to the market that meet the stringent requirements of AI workloads, and as data centers reimagine their architectures for AI, testing is becoming ever more critical,” says Updyke. “A top priority for the Spirent team has been to develop a dedicated test solution capable of emulating realistic xPU workloads and AI traffic patterns with ease. Our new solution will enable engineers to test their Ethernet fabric without having to go to the expense of building a whole new lab of costly xPU servers and configure test cases to generate AI workloads using these real servers.”
Spirent’s new AI test solution runs on the A1 400G platform and is the first hardware that can emulate the high density 400G xPU workloads for AI environments. It enables customers to test their Ethernet fabrics in their existing test environments without having to go to the considerable expense of building new xPU server equipped labs. Utilizing the RoCEv2 protocol, the Spirent platform is designed for ease of use, is straightforward to configure, and provides repeatable and consistent results to reduce the complexity of testing AI use cases. It also boasts the credentials of a true multi-purpose platform that can test both AI and routing/switching use cases concurrently.
“We are proud to be first to market with our AI workload emulation platform and as the AI landscape continues to develop and data center architecture evolves to cater to AI/ML workloads, we are committed to working with our customers to develop the very best solutions to serve their needs,” said Updyke.
For more information, please contact CN Rood.
