KCNA ยท Cloud native principles and patterns
Serverless, FaaS and event-driven scaling
FaaS vs containers, scale to zero, events and brokers, Knative and KEDA role.
Key points
- Serverless means you deploy code and the platform runs and scales it, often down to zero when idle.
- FaaS runs short functions per event; containers run long-lived processes you package yourself.
- Event-driven systems react to events delivered through brokers.
- On Kubernetes, Knative adds serverless serving and eventing, and KEDA scales workloads from event sources such as queue length.
Source: https://keda.sh/docs/latest/concepts/
Practice questions
What does scale to zero mean?
- Idle workloads run with zero replicas and start again on demand
- The app is deleted after each request
- CPU limits are set to zero
- Logs are cleared daily
Show answer
Answer: A. Idle workloads run with zero replicas and start again on demand
With no traffic or events, the platform reduces replicas to zero and starts them again on demand. It does not delete the application definition.
Which are typical traits of serverless / FaaS platforms?
- Automatic scaling with demand
- Billing tied to actual usage
- Functions triggered by events
- Reliable local disk state between invocations
- You patch the underlying servers yourself
Show answer
Answer: A. Automatic scaling with demand, B. Billing tied to actual usage, C. Functions triggered by events
Automatic scaling, per-use billing and event triggers are typical. Persistent local state and manual server patching are not.
True or false: serverless means there are no servers involved.
Show answer
Answer: False
False. Servers still exist; the platform provider manages them so you do not.
Practice 8 questions on this topic in Bitzu
Short lessons, instant explanations and spaced review of what you get wrong. Free on iOS and Android.
