The internet is a tool that many people rely on to support themselves in today’s world. A recent 4-hour outage at Amazon illustrated how interconnected the internet can be, as over a hundred thousand other websites were affected by the temporary outage.

You might think that when one website fails, the rest of the internet wouldn’t be affected. Due to an array of reasons, this isn’t true at all, and the failure of one website can be devastating on the entire internet.

Since the Amazon Web Service is the largest provider of internet computing services, over 100,000 websites rely on their servers to help them avoid investing in their own. As a result, websites like Spotify, Netflix, Instagram, Vine, and much more experienced problems caused by recent failures of the Amazon cloud service.

A cloud analyst was reported to say that the outage was “Either a bug in the code or human error. Right now we don’t know what it was.” However, since websites can also back up files locally, it wasn’t necessarily a panic for many of the websites affected by the outage.

Other sites like the Arlington County of Virginia Website did not have protections against this, and went out altogether. Perhaps this can be a lesson for the future that all files should be backed up locally if you have a website using the Amazon Web Service.

This isn’t the first time this sort of event has happened in recent history. In October 2016, a cyberattack caused Amazon Web Services to experience connectivity issues, resulting in a similar occurrence. Then in 2015, Amazon Web Services went down for five hours. It seems that these outages are starting to become a routine occurrence, with an array of different causes.

With increased centralization of the internet beginning to occur, this sort of problem could have devastating effects on an internet which has a centralized server. It would be then possible for the entire internet to be affected by the failures of a centralized server, making this Amazon outage look miniscule in comparison.