Over the week of June 14, a lot of applications and websites experienced outages, creating problems for many companies until it was resolved. Not only did these outages create business continuity issues, they’ve also exposed just how vulnerable the Internet is to these kinds of issues. Let’s take a look at what happened.
To begin, let’s review the situation at hand.
Fastly provides a content delivery network for numerous influential websites. Basically, they support a network of duplicate servers across 26 countries around the world so that websites can store data and content on them. The closer the user is to where the data is being stored, the faster it can be accessed—which is exactly why news sources use Fastly to ensure users don’t have to wait before they get their news.
However, on Tuesday, June 10th, a large number of websites and applications went down, many of which ranking as some of the most popular on the Internet.
While this outage was relatively quick to be resolved, it still caused significant business interruptions and cost countless millions of dollars. Pretty impressive… particularly when you consider that these damages were the result of a 75 percent decrease in traffic for approximately one hour. Additionally, it wasn’t as though these affected websites could switch to another service on a dime without some proactive preparations (which would do little to simplify the complicated process of doing so).
Let’s consider this scenario as indicative of a concerning factor in how the Internet is now constructed.
While a content delivery network helps to support efficient content delivery (as it should), there are a few potential drawbacks that a CDN presents… despite their growing popularity online. First off, because the CDN draws from a central server, any issues in this server could easily render this content inaccessible. The same could be said of any cybersecurity issues that impact the central server.
CDNs offer considerable speed boosts and cost savings as compared to the undersea infrastructure that would otherwise be used to distribute content, but at the cost of vulnerability to any issues in the system’s backend processes.
Furthermore, because a lot of the Internet’s software specifications are so open, a simple issue could impact millions or even billions of devices. Since it is far more often the case that the issue is a small one, tools like machine learning are now being used to troubleshoot and identify the root causes of Internet outages.
Of course, not all businesses make use of a CDN. Many are now using cloud computing services to support their needs, but even these aren’t immune to issues. Some of the largest cloud providers—Amazon Web Services, Microsoft, and Google—have been known to experience outages from time to time, creating challenges for the businesses that rely on them.
This is not to say that cloud services are useless… quite the opposite. However, it is important that you understand the risks involved in all of your options. Graemouse Technologies can assist you in identifying your needs and customizing your IT to meet them.
To find out how we can make your IT more resilient and effective for your business’ processes, reach out to us at 253.777.0763.
About the author
CEO
Comments