Glossary term
What is Exponential backoff? Meaning for developers
In System Design
Waiting longer after each failure. Example: 100 ms, 200 ms, 400 ms….
Learn it properly · System DesignLesson 41: Timeouts, Retries & Backoff with JitterOne slow dependency with no timeout, plus clients that retry instantly, can take down an entire site. How to set timeouts from real latency, propagate deadlines, retry only what's safe, avoid retry amplification, and why backoff needs jitter.In Practical Design Problems
Double the wait between retries (1 s, 2 s, 4 s…) up to a limit. A struggling service gets time to recover.
Learn it properly · Practical Design ProblemsLesson 7: Design an LRU Cache, Idempotent Payments and Other Classic Mini-ProblemsShort worked designs: an LRU cache, idempotency keys, a job queue with retries and backoff, and feature flags.In API Design
Waiting longer after each failed retry (1 s, 2 s, 4 s) with a little random time added so clients do not retry together.
Learn it properly · API DesignLesson 1: HTTP Methods, Status Codes and Idempotency in PracticeGET, POST, PUT, PATCH and DELETE used correctly, the status codes that matter, and why idempotency decides whether a retry is safe.In Observability & SRE
Waiting longer between each retry, usually doubling the wait up to a cap, so a struggling service gets time to recover.
Learn it properly · Observability & SRELesson 7: Reliability Patterns: Timeouts, Retries, Circuit Breakers and Graceful DegradationFailure modes, timeouts and retries with backoff, circuit breakers, bulkheads, load shedding, tested with fault injection.