I posted these thoughts to LinkedIn on Monday. I had intended to send them out by email on Monday also, but ironically our email list provider (Drip) was out of service due to the AWS outage…
I want to take a moment to give kudos to all the voice providers (and other cloud applications) that were NOT impacted by the AWS outage.
Maybe that’s because you are hosting your own service and have a reliable, resilient network.
Maybe that’s because you are using the cloud, but have a network design that can cope with a provider going out of service.
Maybe you happen to be using another public cloud provider and you just got lucky.
Whatever your situation – use today as an opportunity to circle the wagons.
Let’s understand how your service would have been impacted if something like the AWS outage had happened to you. If you dodged this bullet, then let’s get ready for the next one, and get some body armor in place before it happens to you.
- You could start by looking at the overall architecture of your voice and/or IP networks.
- You could then perform a theoretical “what if” review of the network, asking what should happen if any one element (or IP connection, or cloud provider) were to fail – this helps to identify any risks inherent in the design of the network.
- Better yet, you could design some real-life resilience tests of your network. I’m sure Amazon thought their design was just fine – so finding ways to test the implementation in real-life is a great way to increase your confidence.
No network or product is perfect – and it’s not possible to test every scenario – but let’s use this opportunity to look a little deeper, and build a more resilient service for our subscribers.
If you need some help working through this process, to improve the resilience in your network, please don’t hesitate to reach out.
