As the quality and complexity of our online world has increased, so has the number of people who traverse its digital pathways daily. This new reality, where we have multiple connected devices—everything from computers to phones and even our watches and refrigerators—means that the highways that were once able to provide high speed internet for us are no longer as reliable.
Now, not only are our digital pathways congested, but the quality of content, and its size, are adding an extra strain on an already burdened system. High-quality video streaming, live events, online gaming, and even business networks, put additional stress on a network that was designed with a much smaller bandwidth in mind. The problem is that while our use of the internet has evolved, the infrastructure supporting it has not. According to William Erbey, a serial entrepreneur who has founded 6 multi-billion corporations with a background in informatics, most people view the solution as a hardware problem when in actuality it is a math and software problem.
The difficulty is already becoming apparent and is set to get worse as more people stream greater amounts of high-quality content from all around the world. Until now, the solution has simply been to add more hardware to match the demand, but this linear way of thinking will eventually become unsustainable. Instead, the solution must come from elsewhere, and for many in the industry, artificial intelligence may be the path for overcoming this mounting challenge.
Doing the Same Thing Over and Over Again
Internet congestion is not a new problem. In 2013, the Harvard Business Review noted that congestion was already an issue due to the web’s “all-you-can-eat” model—where open access is a given and users can utilize whatever bandwidth they like—was essentially unsustainable. The idea that simply adding more hardware to provide more bandwidth would fix the issue missed a larger point.
The trouble lies in the fact that the way the internet’s communication works today is largely decentralized and random. Bill Erbey believes the problem is that most companies simply “send out their content, and hope that it arrives at its destination on time.” He adds that “at its core, the current model is based on ideas that are 30 years old, and painfully out of date for today’s reality.”
According to Erbey, this lack of centralized control lies at the heart of the problem. Instead of an organized highway, where data is sent over pathways that are uncongested, traffic is routed over the shortest path regardless of the excess capacity present in the routers along the way. AI can provide a smart remedy to resolve this issue.
How AI Transforms the Solution
Currently, the internet’s routers which direct traffic operate largely independent of each other. Erbey once again notes that “this disorganization creates congestion. Instead of cooperating to create better pathways, traffic hubs operate without considering their counterparts, leading to overlap, waste and heavier congestion.”
Today, some of the most data-heavy online activities account for a large percentage of our usage. Video accounts for 58%; BitTorrent accounts for 22%; and gaming accounts for nearly 8%. These trends are only set to expand, with video becoming increasingly popular, and gaming turning into a mainstream activity worldwide. As such, the answers to congestion must evolve with them. In working with System 73, Erbey and his team have built a solution that utilizes AI to analyze and better distribute traffic, potentially reducing congestion significantly.
The idea is to use AI augmented tree-based peer-to-peer networks that support smarter traffic routing and reduce strain at peak times without sacrificing quality. As opposed to the diminishing returns of hardware-based solutions, AI augmented tree-based networks can scale exponentially. Instead of randomly routing content and hoping for the best, the technology can predict and forecast traffic to make better decisions about the best pathways for delivery. AI can pinpoint problems within networks and across delivery pathways to both fix and prevent them from unfolding.
Congestion Could Be a Thing of the Past with AI
It’s still early, and while AI has already been hailed as a game-changing technology, it must be widely adopted to reach its full potential as a congestion alleviator.
Erbey notes that “AI is still evolving, but we’ve seen measurable results in a relatively short period of time. We need to abandon the idea that simply throwing more money at a problem will fix it.”
The definition of insanity is doing the activity repeatedly and hoping for different results. Indeed, hardware can only take us so far if the systems we build continue to do the same thing they always have. Instead, AI and other emerging technology should be embraced to find creative and sustainable solutions to problems that will only grow worse as our lives continue to migrate online.