Timeouts, Retries, Circuit Breakers & Bulkheads
How slow dependencies take down healthy services, and the resilience toolkit (deadlines, jittered retries, retry budgets, breakers, bulkheads, load shedding, hedging) that stops it.
About 95 minutes · 16 exam questions · 9 interview questions · 2 code labs
What you will learn
- The failure model: slow is worse than down
- Timeouts, timeout budgets & deadline propagation
- Cascading failure in depth: how a slow downstream exhausts upstream pools
- Retries with exponential backoff and jitter
- Retry storms, amplification math & retry budgets
- Circuit breakers
- Bulkheads: isolating resource pools
- Load shedding & admission control
- Graceful degradation & fallbacks
- Health checks vs heartbeats
- Hedged requests & tail latency
- Scenario playbook: slow, down, erroring, storms, cascades