CDNs don’t fail very often; when they do, the impact can be global

Content delivery networks, or CDNs, normally, don’t make it to popular Internet jargon until they stop delivering.
An across-the-world outage that lasted about an hour hit thousands of government, news, and social media websites on June 8. It took down sites including the New York Times, Bloomberg News, Reddit Inc, and even the UK government and the White House.
High-traffic websites including, Amazon, CNN, PayPal, and Spotify were affected too, according to tracker Downdetector.com.
At the root of the problem was Fastly, a popular CDN provider. It attributed the incident to “a service configuration that triggered disruptions across our POPs (points of presence) globally.”
A CDN is a kind of bridge between a website or app and a user, helping to push data quickly around the Internet on behalf of some of the most popular online companies.
CDN providers do that by hosting multiple servers and directing people who call up a web page to the nearest one geographically, rather than sending them back to the
origin. This allows for faster and more reliable delivery of web content.
Such networks are popular with large, high-traffic websites and those offering big files to download.
Along with speed and the greater ease of serving more customers simultaneously, CDNs are better placed to deliver content such as high-resolution video without disruption.
The flip side is they are costly, do not have servers everywhere and, as evidenced by the latest episode, mean that companies are putting the fate of their websites in the hands of an outside party.
For sure, CDNs don’t fail very often. When they do, it can be spectacular. Technically, these issues are also hard to prevent, and often happen when companies need to update their systems.
There is also no evidence to suggest Fastly’s issues were the result of a malicious cyberattack. By contrast, all website system administrators know that network outages and downtime can happen, no matter the size of their hosting platform.
But the fallout would not have been quite rosy for Fastly’s customers. At Amazon alone, the outage could have lost the company $32mn in sales, according to a calculation by the SEO agency Reboot.
“This is what happens when half of the Internet relies on giants like Amazon, Google, and Fastly for all of its servers and web services,” said Gaz Jones, technical director of digital agency Think3. “The entire Internet has become dangerously geared on just a few players.”
When Amazon’s cloud computing unit Amazon Web Services encountered an issue in 2017, some of the world’s biggest websites went offline for several hours across the entire US East Coast.
Just as the rise of the Big Tech has created new challenges for the competition watchdogs, the latest downtime is another caveat about the increased concentration of crucial internet infrastructure among a relatively small number of companies.
In a wider sense, the outage is also a grim reminder of how even a short-lived technical glitch can have huge ramification for global Internet traffic.
Related Story