Refactoring to serverless
summary
Argues that replacing custom code with managed services cuts application code and ops work, using AWS CDK to turn architecture into code (example: four CDK lines for an NLB+Fargate service vs ~450 CloudFormation lines). Shows concrete refactorings: swap SQS calls for Lambda destinations to shrink Lambda runtime and costs, move choreography to Step Functions (200+ service integrations) for transactional integrity, and use EventBridge Pipes with DynamoDB Streams to eliminate in-Lambda event wiring at the cost of higher latency and extra per‑million‑event fees.
AI-written summary · about
chapters (13)
- 0:00Introduction to serverless
- 1:31The role of automation
- 2:27Infrastructure as code
- 8:29Defining serverless refactoring
- 8:51Lambda destination pattern
- 12:47Choreography vs orchestration
- 16:31Database integration patterns
- 18:45Operational trade-offs
- 20:32Motivations for refactoring
- 26:44Nuances in distributed systems
- 27:22Measuring impact on performance
- 32:51Continuous refactoring culture
- 37:02Conclusion and resources
related talks
40mRefactoring to serverless
30mEvolving Serverless Architectures
23mSimplifying design complexities using Serverless Architecture on AWS
36mTransforming SaaS with Serverless Architectures
39mBuilding Highly Scalable Microservices with Serverless
49mMeasuring and increasing developer productivity with the help of Serverless
34mDesigning Data Intensive Applications In Serverless Architecture
54mYou should be building Serverless Microservices