Micro-Frontends in AWS
summary
Micro-frontends map to business subdomains, enforce team ownership, and minimize shared libraries so deployments don’t require cross-team coordination; implementation uses S3 + CloudFront + Lambda@Edge for CDN and edge routing, Fargate for streaming server-side rendering, Step Functions + DynamoDB for on-demand rendering and cache lookups, and Lambdas or container endpoints for each micro-front end. Hydration uses tiny client bundles (example: preact 3KB hydration vs ~32KB), events use a window event bus, and Step Functions caching produces double-digit ms responses in the demo.
AI-written summary · about
chapters (17)
- 0:00Micro-frontends in AWS
- 1:26Defining micro-frontends
- 3:56Independent teams and code sharing
- 8:14Decision framework for architecture
- 11:02Client-side rendering in AWS
- 15:39Server-side rendering challenges
- 24:42Live demo: e-commerce page
- 32:19Step Functions for composition
- 36:16Event-driven communication
- 41:14Infrastructure as code
- 42:59Caching strategies and UI
- 45:25Discoverability and canary deployments
- 46:28Automation and observability
- 48:57Governance and team coordination
- 50:04Key takeaways
- 51:46Q&A: Scaling micro-frontends
- 56:23Q&A: Downsides and trade-offs
related talks
20mTransform Your Web App: Microfrontends & AWS Serverless
20mBuilding Microfrontends on AWS
37mDemystifying Micro-Frontends
36mMicro-frontends discovery
33mMicro-frontends: a migration story on the Edge
37mScalable Microfrontends & How to Optimize Hydration
57mMicro Frontends – a strive for fully verticalized systems
13mMicro Frontends Without Chaos: Architecture That Survives Scale