Scaling Policies and Thresholds Quiz

What you'll be asked

1. What is the primary purpose of defining a 'cooldown period' in an auto-scaling policy?
2. If your application experiences predictable traffic spikes every Monday at 9:00 AM, which scaling policy is most cost-effective and efficient?
3. When configuring a target tracking policy based on 'Request Count per Target', what happens if the actual request count exceeds the defined target value?
4. You notice your application experiences 'flapping'—where the system scales out and in repeatedly within a short timeframe. Which configuration change would best mitigate this?
5. In a distributed system using predictive scaling, the algorithm suggests a capacity of 10 nodes for an upcoming peak, but the actual observed CPU utilization is very low. Why might the algorithm still maintain 10 nodes?

Answers and explanations come after you submit.