Intelligent Health Checks: one-click observability for reliability tests

Intelligent Health Checks, introduced by Gremlin, automate the process of reliability testing and observability by enabling engineering teams to easily monitor and test their services without the need for third-party tools. This feature automatically configures Health Checks based on the metrics of error rate, latency, and request rate—three of the four Google Site Reliability Engineering handbook's Golden Signals—by observing a service's metrics in AWS CloudWatch and setting reasonable failure thresholds. Enabled with a simple checkbox within Gremlin for AWS, Intelligent Health Checks integrate with AWS services, such as Elastic Load Balancers, to assess the health of services during tests, halting them if thresholds are exceeded. Gremlin's approach allows teams to balance reliability with other priorities like feature development and incident response, providing them with a tool to find and fix availability risks before they affect users.