SAA-C03 Practice Question 30: ASG Scheduled Scaling Cost
AWS Certified Solutions Architect – Associate (SAA-C03) · Test 3 · question 10 of 10
An internal claims-processing application runs on a fixed Auto Scaling group of 20 On-Demand EC2 instances, 24 hours a day. Monitoring over the past year shows a very regular pattern: all 20 are needed on weekdays from 08:00 to 18:00 local time, and 4 are enough at all other times.
New instances need about 15 minutes to warm their caches, and the full fleet must be serving by 08:00 when staff log in. The application keeps work in progress in memory, so its instances must not be interrupted. Which change reduces cost the most while meeting these requirements?
Show answer and explanation
Answer:
Scheduled scaling is designed for load changes you can predict. You create scheduled actions that set the group's desired capacity, and optionally its minimum and maximum, at specific times. Using a cron expression and an IANA time zone, which adjusts for daylight saving time, the actions can repeat every weekday. Starting the scale-out at 07:30 leaves time for the 15-minute warm-up, so the fleet is ready at 08:00. Dropping to 4 instances outside business hours cuts the instance-hours billed from 480 a day to 256 on weekdays and 96 on weekend days. You can still add a dynamic policy on top for unexpected spikes.
Option A reacts only after load arrives. At 08:00 CPU climbs, the policy starts launching instances, and staff wait out the 15-minute warm-up on an undersized fleet. The requirement that the fleet be serving at 08:00 rules this out.
Option B discounts the wrong thing. Committing to 20 instances around the clock pays for 16 instances that sit idle for more than half of every week. Shrinking the fleet on a schedule removes that waste entirely.
Option C conflicts with the stated constraint. Spot Instances can be interrupted when EC2 needs the capacity back, and the application's in-memory work in progress can't tolerate that.