Why do you need distributed uptime monitoring?
The world is connected and geographical distance is not a problem thanks to the Internet. Websites, blogs, and apps are accessed by millions and millions of people around the world. You can buy a watch from a store in your city or thousands of miles away. And the simple fact is: if you have a website, server or app, you must monitor it!
But why do you need distributed uptime monitoring?
Now we will bring the 2 big reasons you must consider distributed uptime monitoring for your websites, servers, and apps, no matter the size of your company or the target audience:
#1 — Understand resource allocation needs
According to Jakob Nielsen, users in 1993 didn’t wait more than 10 seconds for website response. According to Joshua Bixby, the average expectation of a shopper in 2010 was to wait no more than 2 seconds for the response from an e-commerce website. As the technology is developed, the expectations on responsiveness get higher and higher.
Responsiveness matters!
Distributed monitoring systems can be useful to provide a better understanding of performance issues related to specific geographic areas. If you have a monitor producing high response times from the UK agent, it’s a huge indication that your Britain users are facing the same performance issues.
You can use these insights to identify how to better allocate resources, as well as to decide to use a replication or CDN service. Reallocating a website or replicating a server to a data center that is “closer” to your users will definitely improve their experience.
#2 — Figure out broken routes and geographic unavailability
Monitoring your websites or servers from your office’s network is out of the question unless you have only internal access to them. Indeed, a better solution is to have a computer or system outside the network to monitor from an external source. The monitoring would identify any unavailability, caused either by internal or external factors.
That is a good solution! However, you can still have false negatives!
and let’s find out why…
In fact, 2 different endpoints on the internet (e.g.: a user and your website) can be connected through different paths. It depends on the distance between them and the complexity of the infrastructure in between. In addition, internet routing is, most of the time, defined dynamically. That means there is no guarantee that a user will always connect to your website through the same routes.
A single external monitoring source can lead to false negatives, due to errors in specific routing paths. And the best way to mitigate this risk is to rely on distributed monitoring systems. They can double-check from other sources when a failure is recognized. It will not totally remove the possibility of false negatives, but will drastically decrease it. And it will provide information on broken routes, and then advise you on the group of users that might be facing issues to access your website or server.
Distributed systems will produce more reliable uptime/downtime insights. Thus, relying on them to notify you when your website or server goes down is indeed the best option.
