Investments in AI, machine learning and network capacity have helped Comcast (News - Alert) meet unprecedented internet demands during the COVID-19 pandemic. In a recent interview, the company explained that its suite of AI and machine learning tools provides visibility of its entire network, enabling Comcast to quickly add capacity and troubleshoot issues before performance is impacted.
Comcast recently told VentureBeat that its network traffic rose 32 percent in March at the onset of the coronavirus pandemic. It also experienced a 36-percent increase in mobile data use over Wi-Fi on Xfinity mobile during that time. The company's network is accessible to more than 59 million homes in the U.S. through 800,000 miles of cable.
“With COVID-19, we obviously saw a massive surge in the network, and looking back in retrospect the network was highly reliable,” said Jan Hofmeyr, chief network officer at the Comcast Technology Center in Philadelphia. “We were able to respond quickly as we saw the spike in traffic. We were able to add capacity without having to take the network down. It was designed for that.”
The company credits its Comcast Octave software with managing the increased traffic and complexities. Developed internally by Comcast engineers, the AI platform checks more than 4,000 telemetry data points on more than 50 million modems throughout the network every 20 minutes. The software looks for variables like power levels, external network noise and other issues that can negatively impact performance.
Octave detects when modems aren't using all the available bandwidth as efficiently as possible, then automatically adjusts them. This enables increases in speed, efficiency and overall network capacity. The company also used machine learning to put the data collected from this process into algorithmic solutions. This helps Comcast predict where interference may disrupt networks or trouble points may appear.
The AI solution had only been rolled out to part of Comcast's network at the onset of the pandemic. The company then put together a team of dedicated engineers to speed up the deployment process and make use of the technology throughout its network. Comcast said customers enjoyed a nearly 36-percent increase in capacity as a result of its efforts.
“We have to turn the knobs so that we optimize delivery to your house, which would not be the same as the delivery to my home,” said Elad Nafshi, senior vice president for next-generation access networks at Comcast Xfinity. “We provide you with much more reliable service by detecting the patterns that lead up to breakage and then have the network self-heal based on those patterns. We’re making that completely transparent to the customer. The network can self-heal autonomously in a self-feedback loop. It’s a seamless platform for the customer.”
Octave is used in tandem with Comcast's Smart Network Platform, also developed in-house. The suite of software tools automates core network functions, which cuts down on the number of outages and shortens their duration. The network is also more secure and has built-in redundancies, enabling it to hold up to issues like an accidental cutting of a fiber-optic cable.
Comcast also uses the NetIQ (News - Alert) machine learning tool to continuously scan its core network, collecting thousands of measurements per hour. This enables the company to instantly see outages and reduce the time it takes to remediate them.
The use of AI and machine learning to improve network performance extends well beyond service providers. Enterprises are increasingly adopting AI for network performance management and optimization as well as a host of other uses. To learn more about the role of AI in the enterprise network and beyond, TMC (News - Alert) is hosting a Future of Work Expo from June 22-25, 2021 at the Miami Beach Convention Center. The event will examine how AI and supercomputing are changing the workforce and the global information economy as we know it, with applications in the data center, contact center, customer service, sales and marketing and more.