Overloading a software system occurs more often than expected, and the effects are difficult to deal with, including real-time web services halting and asynchronous systems building up backlogs. In this talk, we cover what AWS does to build reliable and resilient services, including avoiding modes and overload, performing bounded work, throttling at multiple layers, guarding concurrency, sending idempotent requests, applying backpressure and fairness in queueing, and performing shuffle sharding. We also discuss how separating concerns through service-oriented architectures helps reduce blast radius. As we explore these patterns, we discuss how they’re embedded into the DNA of the AWS services that you use to build and operate serverless applications that are resilient to failure. We also discuss a number of AWS services, including AWS Lambda, Amazon API Gateway, AWS WAF, Amazon CloudWatch, and AWS X-Ray.