From Backend Powerhouses to Web-Friendly APIs: Unlocking
summary
Shows a practical recipe: replace microservice REST glue with gRPC over HTTP/2 with protobuf for interservice calls (multiplexing, bidirectional streaming) and move client-facing stitching to a GraphQL BFF cutting 3–4x API calls so mobile requests only fetch needed fields. Add CDN-edge caching with event-driven invalidation to avoid redundant trips to databases. Results: gRPC yields ~75% serialization efficiency vs JSON, connected-vehicle latency drops from ~2000ms to 630ms (75th pct), mobility pages fall to ~230ms.
AI-written summary · about
chapters (21)
- 0:00Website speed for business
- 1:10Web vitals measure performance
- 2:34API performance challenges
- 3:39REST limits and HTTP 1.1
- 4:45gRPC outperforms REST
- 6:10gRPC efficiency and scalability
- 8:16gRPC architecture benefits
- 9:41gRPC vs REST performance
- 10:02Mobile apps need efficient data
- 10:45GraphQL optimizes frontend queries
- 11:49GraphQL in BFF layer
- 12:31GraphQL for mobile and web
- 13:57Caching strategies improve performance
- 14:39Optimal caching placement
- 16:25CDN caching for web apps
- 17:07Combining gRPC, GraphQL and caching
- 17:30API response time improvements
- 18:32API and web performance factors
- 21:40Kubernetes networking pitfalls
- 22:42API design trade-offs
- 23:25Q&A
related talks
59mGraphQL, gRPC or REST? Resolving the API Developer's Dilemma
51mGraphQL, gRPC or REST? Resolving the API Developer's Dilemma
06mBeyond REST: Exploring Modern API Paradigms
17mBoost API Performance: gRPC vs REST vs Async
28mRequiem for APIs - The Evolution of Web Development
26mDemystifying Protocol Buffers and gRPC
50mgRPC Web: It’s All About Communication
42mREST API vs gRPC, Which One Should You Use in Breaking a Monolith