Mike Jack, director of product marketing at Viavi, told RCR that the challenge is not only validating individual links but ensuring interoperability and performance across a rapidly evolving ecosystem of devices, optics, and architectures
In sum – what to know
Speeds raise complexity – Higher data rates increase testing challenges around signal integrity, noise, loss and interference.
Interoperability matters – Validation must cover devices, optics and architectures across a rapidly evolving ecosystem as networks move toward terabit speeds.
Power and heat – Higher-speed, denser deployments require validation of power consumption, thermal behavior, airflow and system stability.
The move toward higher-speed optical interconnects such as 800G and 1.6T is creating testing challenges at both the physical and system levels in AI data-center environments, according to Mike Jack, director of product marketing at Viavi, in an interview with RCR Wireless News.
“At the physical layer, higher speeds push signal integrity to its limits, requiring more advanced instrumentation and validation techniques. These higher data rates increase susceptibility to noise, loss, and interference, making accurate testing more complex and demanding,” the Viavi executive said.
At the system level, Jack said the challenge is not only validating individual links but ensuring interoperability and performance across a rapidly evolving ecosystem of devices, optics and architectures.
Jack said industry demonstrations, such as those at OFC 2026, highlight the importance of multi-vendor interoperability testing as networks scale toward terabit speeds. At the same time, traditional throughput testing is no longer sufficient, with AI workloads requiring validation of synchronized performance, low latency, and reliability across the entire fabric.
Jack also pointed to the preparation for higher speeds such as 3.2T. “This is particularly important as the industry prepares for even higher speeds such as 3.2T, which will further increase system complexity,” he said.
Power efficiency and thermal management are also becoming considerations in AI networking as higher-speed optics and dense deployments increase power consumption and heat generation, the executive said.
Jack said the transition to 800G and 1.6T, along with early preparations for 3.2T, is driving innovation in optical design aimed at improving efficiency, reducing power per bit and enabling higher density. These developments include new form factors, silicon photonics and architectures optimized for cooling and space constraints, according to Jack.
From a validation perspective, testing must extend beyond performance metrics to include power consumption, thermal behavior, and system stability under real operating conditions. High-density environments also require validation of airflow, cooling efficiency, and the interaction between physical infrastructure and network performance.
Jack said maintaining performance within power and thermal limits is becoming an important part of validation as AI infrastructures scale. “As AI infrastructures continue to scale, ensuring that systems can maintain performance within power and thermal limits is becoming as critical as validating bandwidth and latency,” he said.
