aiengineering.guideaiengineering.guide

Glossary

rate limiting

Capping how many requests a caller may make in a window, to protect a shared resource.

Precisely, a policy that bounds the number of admitted requests per caller over a time window, rejecting or delaying the excess. Confusions: rate limiting is not the same as load shedding (which drops work under overload regardless of caller) or quota (which is billed usage over a long period).

Also: rate limit, rate-limit

Enter to go · Esc to close