On June 23, 2026, VIAVI Solutions Inc. (NASDAQ: VIAV) announced a significant advancement in the field of network technology with the launch of the industry's first Ultra Ethernet Transport (UET) validation solution tailored for AI fabrics. This innovative solution is designed to empower hyperscalers, cloud providers, and network equipment manufacturers by introducing powerful traffic generation and analysis capabilities that promise to accelerate the deployment of AI data centers.

2. Accelerating AI Data Center Deployments

The newly developed UET validation solution is a GPU-free, full-fidelity offering integrated into the VIAVI TestCenter platform. It aims to facilitate the swift rollout of next-generation high-speed AI networks by effectively validating both scale-out and scale-up AI back-end networks. This capability is built upon the Ultra Ethernet Consortium's (UEC) Ultra Ethernet Stack, specifically designed for large-scale AI and High-Performance Computing (HPC) workloads.

Key Features of the UET Validation Solution

The UET validation solution is engineered to ensure that AI fabrics deliver the requisite performance and resiliency without incurring the high costs and complexities associated with dedicated GPU infrastructure. Its capabilities include:

Stateful AI Traffic Pattern Emulation: The solution can replicate realistic AI traffic patterns at scale, including reliable ordered and unordered delivery (ROD/RUD), packet trimming, congestion control, and dynamic multipathing.

Comprehensive AI Workload Emulation: It supports full-fidelity emulation of critical AI workloads, including collective communications (CCL) and flows associated with large language models (LLMs).

Load-Balancing Mechanisms Validation: The platform enables thorough validation of various load-balancing techniques across the AI fabric, such as Equal-Cost Multi-Path (ECMP), packet spraying, and flowlet switching.

3. Insights from VIAVI Leadership

Aniket Khosla, Vice President of Product Management for Optical Transport and High-Speed Ethernet at VIAVI, remarked on the challenges posed by the rapid growth of AI clusters. "AI clusters will soon scale to millions of endpoints, which means relying on physical GPUs alone to validate network behavior is no longer practical," Khosla stated. He emphasized the importance of rigorous testing for the success of Ultra Ethernet, which delivers high throughput for AI and HPC applications.

Additionally, Mahesh Subramaniam, Senior Director of Product Management for AI Data Centers at HPE, highlighted the need for advanced validation techniques as AI data center fabrics grow increasingly complex. He noted that "the VIAVI TestCenter platform provides realistic traffic emulation and the granular visibility needed to validate congestion control and transport performance at scale."

4. Collaboration with Industry Leaders

The launch of the UET validation solution is also a testament to VIAVI's collaboration with industry leaders. The solution has been validated using the Juniper QFX5240 platform and Junos Evolved, which includes advanced software features such as packet trimming. This partnership underscores the role of advanced switching and realistic traffic testing in enabling faster deployment and reliable scaling of Ultra Ethernet in next-generation AI networks.

5. Conclusion

With the introduction of the Ultra Ethernet Transport validation solution, VIAVI Solutions Inc. is poised to revolutionize the landscape of AI data center deployments. By providing a cost-effective and efficient way to validate the performance of AI fabrics, VIAVI is not only addressing the immediate needs of hyperscalers and cloud providers but also setting the stage for the future of high-performance networking in the AI domain.

As organizations continue to seek scalable solutions to meet the demands of AI and HPC workloads, VIAVI's latest offering stands as a pioneering solution that combines innovation with practicality, reinforcing its position as a leader in test and measurement technologies.

Original press release