Comcast recently completed a networking trial with AMD and Classiq that showed the ability to leverage a quantum system to optimize network path resilience in the event of multiple route outages.

The trial was targeted at finding independent backup paths for network sites “when implementing network maintenance and change management.” Specifically, it considered a situation where if a network site was taken offline for maintenance and a second site was to then fail, could an autonomous system tap into a quantum system to reroute network traffic without impacting network quality.

“To achieve this outcome, operators must identify unique backup paths that are fast, resilient to simultaneous link failures, and optimized for the lowest latency delivery, a task that becomes exponentially harder to identify as networks grow,” stated the trial’s task. “Optimization problems in global telecommunications networks represent large combinatorial search spaces that grow exponentially with network size, making them computationally intensive to solve – the perfect challenge for quantum computing.”

AMD provided its Instinct GPUs to support “qubit scale” computational capacity to apply “quantum techniques” that ran alongside classical computing. These techniques were run on “quantum hardware and in accelerated simulation environments … not yet possible through quantum hardware alone.”

Classiq provided quantum software and engineering support for modeling, implementation, and execution of the trial across hardware and simulated environments.

“With the GPU-accelerated simulations, the teams were able to iterate rapidly and validate algorithm behavior, together with runs executed on quantum hardware to assess implementation success,” the group noted, pointing to detailed results of the trial.

Comcast Chief Network Officer Elad Nafshi noted in a statement that the trial was initiated last year, with the “goal of understanding how quantum software and technology could tackle real network challenges. Our results have shown that quantum computing for network optimization isn’t theoretical – it’s practical, scalable, and grounded in the needs of our customers.”

Quantum to manage virtualization

The work builds on Comcast’s long-standing Project Genesis network virtualization efforts. That program was initiated as a way for the cable giant to virtualize network control of its fiber and coax infrastructure.

Nafshi during an investor conference last year explained that the project was on track to hit 70% “execution” by the end of 2025.

“When you look at where we are today, we start by rolling out something called mid-split over this virtualized architecture,” Nafshi told the investor audience in early 2025. “And what mid-split enables us to do is offer multigig services first and foremost, but it also enables us to change the parameters in terms of the speed symmetry and offer a lot more symmetric speeds. It puts all of the architecture that is necessary to have that seamless automation, to have that AI-driven operations, to have the real-time telemetry and visibility into each of our customers' homes, and be able to optimize that with AI that we've built and architected in order to truly optimize the customer experience down to the customer's home.”

Nafshi added that this architecture was superior to what Comcast’s fiber-only rivals could accomplish.

Unlike the fiber competition, I could basically have glass pipes that lead to the home, and if they break, they break. For us, we have embedded AI in the network, literally feet away from the customer, with real-time pattern detection capabilities that says what is going on with your home. And we're able to detect that in real time, and we're able to action on that in real time,” Nafshi said. “And so this is a transformation not just from the serviceability and the speed, etc., but also with what we are now able to offer in terms of reliability and in terms of scale.”