FaultTolerance
Continuing to operate correctly when a component fails, vs merely recovering after (that's DR).
Why we need this / what value this brings
The difference between a blip and an outage is often whether individual components fail independently or take the whole system down with them.
When to use this
Whenever a new external dependency or failure-prone component is added.
How to use or implement this
Apply the patterns in Idempotency/ (circuit breakers, retries, bulkheads) at each integration point that can fail.
Research questions
- Where does fault tolerance overlap with Idempotency/CircuitBreakers — same concern, different altitude?
Empty folder — drop notes, links, and findings here as you research.