Cloud architecture patterns, how to implement them in a serverless way
summary
Practical serverless recipes: implement storage first by putting API Gateway → SQS/Kinesis before processing (the uploader describes a real system that collected phone data and survived spikes where ~95% install simultaneously), use Lambda, Step Functions, EventBridge, DynamoDB and CloudWatch to add a circuit breaker and retries with jitter, prefer JSON with an envelope (metadata/data) pattern (credited to Sheen Brisals) for tracing and retry counts, and adopt saga choreography, data enrichment, priority queues and queue-based load leveling as proven, workable patterns.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 3:00Event-driven systems
- 4:07Orchestration vs choreography
- 5:57Envelope wrapper pattern
- 6:59Commands vs events in JSON
- 8:24Resiliency with SQS and Kinesis
- 11:59Circuit breaker pattern
- 18:07Retry with backoff and jitter
- 26:54Saga orchestration pattern
- 31:02Data enricher pattern
- 34:13Priority queue pattern
- 37:26Queue-based load leveling
- 39:53Summary
related talks
49mServerless Architectural Patterns and Best Practices
51mServerless architectural patterns
23mSimplifying design complexities using Serverless Architecture on AWS
45mPatterns & practices for developing enterprise-scale serverless applications
42mNo Server Is Better Than No Server: Building Blocks for Serverless Architectures
48mServerless design patterns
01h 02mServerless Architecture Patterns
50mServerless Architecture Patterns and Best Practices