Load Shedding

Definitions

1. A resilience strategy where a system intentionally drops or rejects excess requests during overload to protect core functionality. Unlike rate limiting (per-client), load shedding is global — the system monitors its own health and starts rejecting requests before it crashes. Returns 503 Service Unavailable.

   In plain English: A system intentionally refusing some requests when it's overloaded, to prevent a total crash.

   Example: "When CPU hits 90%, load shedding kicks in and drops non-critical analytics requests to keep the checkout flow alive."

Related Terms

• Circuit Breaker
• Backoff
• Connection Pooling
• Islands Architecture
• Debounce
• Throttle
• Thread Pool
• Backpressure