Learn from AWS Fault Injection Service team’s approach to Game Days

The article discusses how the AWS Fault Injection Service (FIS) team conducts game day exercises to improve operational resilience and test system reliability.

• Game days are structured exercises that simulate failure scenarios in pre-production environments
• Key objectives include testing system responses, training on-call operators, and validating monitoring systems
• The team follows a comprehensive framework with detailed preparation, execution, and post-event analysis
• Game days help identify gaps in runbooks, improve incident response times, and build operator confidence
• Best practices include starting simple, running exercises frequently, and focusing on business-critical services

The approach enables continuous improvement of operational capabilities, helping teams build more resilient systems and more confident operators.