Rate limiting
Rate limiting from both sides of the wire: why APIs limit requests, how limits are expressed, the 429 status and Retry-After and RateLimit headers, handling limits as a client with backoff and jitter, quotas and concurrency limits, limiting at Nginx and gateways, and choosing what to key limits on.
CodexBash, , 5 min read
Almost every API you call has rate limits, and almost every API you build needs them. Limits protect servers from overload, keep one heavy user from degrading service for everyone, slow down attackers guessing passwords, and control the cost of expensive operations. Our system design article on rate limiters covered the algorithms and a distributed implementation. This one covers rate limiting as it appears on the network: how limits are communicated, how well-behaved clients handle them, and how to configure them at the edge.
Sign in to read the rest
The whole 5 minute article is free with an account, along with every problem, lesson and tutorial. No card.