Chaos Engineering is the discipline of experimenting with a system with the aim of increasing confidence in its ability to withstand problems in your environment.
Testing in non-productive environments should be performed regularly and should be part of the CI/CD cycle. In production, teams must perform the tests in such a way as not to cause the service to be unavailable.
Test results should be measured and then compared with availability objectives to understand whether the application running in a particular environment is able to achieve the defined objectives.
AWS FIS is a service dedicated to simulating failures in the AWS environment so that testing can be carried out in a controlled manner. AWS FIS simplifies the process of controlling the experimentation of applications in a crash environment. For more information, see: AWS Fault Injection Simulator
References: