Artificial intelligence and machine learning technologies have been integral to solving the challenges of the COVID-19 pandemic. A number of countries around the world have used machine learning to enhance their contact tracing strategies, develop predictive analytics tools to identify future hotspots and come up with new solutions to improve social distancing. Some people were skeptical of these new solutions, but they still proved effective.
The COVID-19 pandemic nearly pushed the Internet to its breaking point. A report back in March showed Internet traffic increased by 70% since pre-pandemic levels.
Many Internet service providers had to find innovative approaches to handle the growing bandwidth and traffic requirements. Comcast and many other leading Internet service providers used artificial intelligence to keep the Internet running efficiently. This included developing new ISP billing system options and other features to streamline their services.
How Comcast and other ISPs use AI to tackle growing traffic requirements
The world was not ready for the challenges that the pandemic created as more people used the Internet on a regular basis. Internet speeds dropped noticeably in many areas, especially those with a high concentration of Internet users. Rural areas with a lot of regular Internet users were some of the worst hit, because they didn’t have the infrastructure to handle the sudden spike in traffic. Traffic spikes also increased among the IoT, as more people recognized its benefits.
Many companies had to decrease their download speeds to reflect growing demands on the resources. ISPs had to play a role in addressing these challenges. Fortunately, new machine learning tools help them improve network speeds, even as traffic demand continue to rise due to a growing number of people working from home.
Tony Werner, the president of technology for Comcast, talked about some of the challenges the company has faced in recent years. The traffic surge brought on by the pandemic was one of these issues, but other factors caused burdensome traffic demands as well.
Werner said that the company developed artificial intelligence algorithms to address some of the most pressing challenges of 2020. Comcast developed a new AI platform to identify anomalies in their networks. These algorithms are also being tweaked to come up with the most appropriate solutions to the various problems the networks encounter.
It’s not just automating what smart engineers can do. It’s going to places where they just couldn’t process that amount of information and come up with solutions quick enough to do what [Octave] does ¦ All you need is a table setting that gets wrong in a router someplace, and all of a sudden traffic very intermittently starts going to the wrong location, he told Fast Company.
Many machine learning applications focus primarily on streamlining processes that would otherwise be handled by engineers and system administrators. The goal of these algorithms is generally to accomplish the same thing as humans in a shorter timeframe.
However, the new AI solution developed by Comcast is considerably more advanced. Werner points out that it is able to conduct very intricate calculations and perform highly detailed analysis that even the most competent engineers would be incapable of completing, regardless of the amount of time available to them.
In the case of the COVID-19 pandemic, these new AI protocols can help identify traffic spikes. They can ensure that networks have sufficient resources to handle them without collapsing under the strain.
The average layperson might think that AI might be unnecessary in these instances. They might think that the best way to prepare for increasing traffic demands is by simply scaling their infrastructure. Although this approach might work to some degree, it has its own limitations. The relationship between network traffic and resource requirements is far from linear. Also, ISPs can encounter a variety of logistical problems, even if they have the requisite infrastructure in place. Adding additional servers and other infrastructure elements to an ISP network might do little good if data cannot easily be transferred across the network.
Artificial intelligence helps tackle the nuanced challenges this presents. ISPs will be able to find new and more effective solutions to adapt to raising traffic demands.
The COVID-19 pandemic has forced ISPs to identify these types of challenges. However, the solutions that they are setting in place will be viable for future traffic searches as well. They will continue to build on their artificial intelligence technology to ensure that they can address the growing problems that they will face in the post-pandemic era.
