A company uses an auto-scaling group to handle variable web traffic. The scaling policy is based on CPU utilization. However, during a flash sale, the application experiences a sudden spike in traffic, and the auto-scaling group takes too long to launch new instances. Which design change would most effectively reduce the time to scale out?
Trap 1: Use larger instance types to handle more traffic
Vertical scaling doesn't improve scaling speed.
Trap 2: Lower the CPU utilization threshold
This triggers scaling earlier but doesn't reduce instance launch time.
Trap 3: Increase the cooldown period
Increasing cooldown delays further scaling actions.
- A
Use larger instance types to handle more traffic
Why it fails: Vertical scaling doesn't improve scaling speed.
- B
Lower the CPU utilization threshold
Why it fails: This triggers scaling earlier but doesn't reduce instance launch time.
- C
Increase the cooldown period
Why it fails: Increasing cooldown delays further scaling actions.
- D
Use scheduled scaling to pre-warm instances before the sale
Proactive scaling reduces response time.