This virtually unknown company just disrupted the entire global Internet Internet

On Tuesday morning (June 8), at around 10:50 a.m. BST and 5:50 a.m. ET, a bunch of hit websites were inaccessible for about an hour. These include Reddit, Amazon, Twitch, CNN, The New York Times, BBC, PayPal, HBO Max, Hulu, PayPal, and even the UK government website.

All of them have only one thing in common: using the service of Fastly, a content delivery network (CDN) provider.

You may not have heard of Fastly, but you probably already interact with it in some way every time you go online. Along with Cloudflare and Akamai, they are the three largest CDN providers in the world. And while Fastly itself is still vague about what caused the worldwide disruption, it also serves as a stark reminder of how connected our internet infrastructure is. with each other and how fragile, especially when it largely depends on the operating processes of a few companies that are still outside the public’s awareness.

Special “goods” Món

To understand in a simple way the role and importance of Fastly, let’s take a minute to learn about the role of CDN in the Internet ecosystem. While many people still consider the Internet to be amorphous, the articles you read, the movies and songs you’re streaming, the photos you post, they all reside on physical servers. And while that content may be primarily hosted on a cloud service provider, there still needs to be a way to get it to people quickly and efficiently.

That’s where CDNs come in. By operating servers around the globe, a CDN can bridge the gap between your smartphone and the Internet experience of your choice. Think of it like how a ball travels in baseball: Instead of trying to throw the ball into his own home field, a player will throw it to a teammate who will throw it to the other player. catch the ball. This is faster and more efficient than throwing the ball straight home.

“Basically, CDNs give really high performance to content, whether it’s streaming video or a web page, or all those little images that appear when you visit an e-commerce site. “, said Angelique Medina, director of product marketing at network monitoring company ThousandEyes. “It allows users to greatly reduce page load time and also allows people to have a really great experience when they are surfing the web.”

Take for example this article you are reading right now. Chances are what you’re reading is a copy of it, cached by something called a “point of presence” (POP), or a server somewhere in your area. The Fastly network map shows that the company operates POPs in at least 58 cities around the world, including many in densely populated areas such as Los Angeles, London, and Singapore. It lists the combined global capacity at a whopping 130 terabits per second.

List of major players affected by the Fastly incident.

And that’s not all! CDNs don’t just store content closer to the devices that want it. The company also helps navigate them on the internet.

“It’s like coordinating the flow of traffic on a major road system , “ says Ramesh Sitaraman, a computer scientist at the University of Massachusetts at Amherst who helped create the first major CDN as an ant. principle architect at Akamai company said. “If some links on the internet fail or become congested, the CDN algorithms quickly find an alternative route to the destination.”

So when a CDN goes down, it can drag parts of the Internet with it.

Exclusive CDN

“We identified a service configuration that was causing disruption across our POPs globally and disabled it,” a Fastly spokesperson said in a statement. “Our global network will be back online.”

Again, it’s unclear exactly what happened at Fastly. The term “service configuration” can mean anything, but the only certainty is that whatever the root cause is, it has a wide-ranging impact. According to Fastly’s crash reporting site, every continent around the globe, apart from Antarctica, felt the impact of the incident.

And even after Fastly has essentially fixed the problem, the company warns that users may still see a lower “cache hit rate” – how often you can find what you’re looking for. already cached in a neighboring server – and “origin load increases”, a concept that refers to the process of returning to the source for items that are not in the cache.

CDNs are typically designed to withstand extreme conditions and unexpected crashes. If one server fails, other servers can take over the load. If the entire data center fails, the transmission pressure can be transferred to other data centers. If everything is working perfectly, then when something goes wrong, the CDN’s recovery mechanisms will ensure that the user almost never notices anything out of the ordinary.

But when something does happen, it’s usually related to a software bug or a configuration error being pushed to multiple servers at once, according to Sitaraman.

Even then, sites and services that use CDNs often have their own fallback measures. The giant Amazon, for example, only takes about 20 minutes to get up and running, as it can redirect traffic to other CDN providers. But for those who rely solely on Fastly or do not have an automated system for troubleshooting, they have to wait.

When a CDN goes down, the impact is far-reaching.

“The outage was the result of a monopoly,” said Roland Dobbins, principal engineer at security firm Netscout. He suggests that every organization with a significant online presence on the Internet should have multiple CDN providers to avoid this type of situation.

However, their options are increasingly limited. Because just as cloud data systems are largely in the hands of Amazon, Google, and Microsoft, the three CDN providers Cloudflare, Akamai and Fastly have dominated the online content stream.

“There are a lot of people focused on using the services of very few service providers,” Medina said. “Whenever one of those three providers has a problem, it’s usually not a long-term problem, but it’s going to have a big impact on the Internet.”

And according to Medina, over time as more and more people have to rely on these handful of companies, the more problems occur, the more connections are missed and the bigger the problem becomes.

Refer to Wired