High Performance APIs through Caching
summary
Argues that caching API responses and event-driven purges let APIs be fast and cheap rather than just “API-first” fluff, using CDNs and headers to control TTLs and browser behavior, tag-based surrogate-key batch purge to invalidate groups, and edge invalidation at 150 ms (95th percentile) on Fastly. Gives real wins: New York Times offloads alerts and saves $25,000 a month, HotelTonight uses geofenced queries and Boots uses GeoIP for smarter routing.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 0:56History of APIs
- 2:12The Web
- 11:13Fastly
- 12:42Fetch
- 14:46Visual Model
- 15:44Surrogate Control
- 16:49Batch Purge
- 18:46Examples
- 19:43Geofencing
related talks
25mCaching strategies on https://dev.to
47mCaching strategies on https://dev.to
52mEdge Side APIs : Fast and Reliable Hypermedia APIs
37mCaching Deep Dive: The Ultimate Way To Speed up Your GraphQL API
30mCaching, Payloads, and Other Dark Arts
42mCaching for Cash
38mThe power of cache in a slow world
40mLiving on the Edge
