Part of the Gremlin Office Hours series: A monthly deep dive with Gremlin experts. To get the most value out of Chaos Engineering and reliability testing, you need a way to observe your service’s behavior. Observability tools offer insight into how your systems are performing, but observability on its own isn’t enough. You need a way to monitor your systems while testing their reliability so you can determine whether your service passed or failed a test.

Unreliable software can have a painful impact on your customers and your business—something we’ve all seen and felt during high-profile outages. And while building on the cloud with AWS unlocks improved scaling and reliability capabilities, the complexity of modern distributed systems can potentially introduce outage-causing reliability risks. How can you be sure your systems are resilient to failure when they’re based on complex architecture, built by hundreds of teams, and are being updated almost constantly?