Building reliable serverless applications - AWS User Group Peru August 15 2020

Presented at AWS User Group Peru August 15th 2020.

@gunnargrosch

Serverless and fully managed services give you high availability and robustness out of the box, but even though every piece of your architecture might be resilient to failure you still need to use well-architected patterns and practices to make your application reliable. In this session we'll dive head first into the world of robustness, reliability and resilience to examine some of the patterns and practices we use to build battle-tested serverless applications. We'll also look at how to verify the output of the system through chaos engineering and the advantages established by CI/CD.

Transcript

18. /login
71. • • • • •
73. • • • • • { "isEnabled": false, "failureMode": "latency", "rate": 1, "minLatency": 100, "maxLatency": 400, "exceptionMsg": "Exception message!", "statusCode": 404, "diskSpace": 100, “denylist": [ "s3.*.amazonaws.com", "dynamodb.*.amazonaws.com" ] } const failureLambda = require('failure-lambda’) exports.handler = failureLambda(async (event, context) => { ... })