The non accepted answer by @Az-fta was the right one. If this is AWS scaling then it will work as the accepted answer.

Proof:

1. There was a discussion in reddit and someone with Proof scenario

2. Another example was from Udemy, Alan Rodrigues course for Azure 104. A Cooldown period of 5 minutes is added and scale-out is set for every 5 minutes.

`Good job!

The first scaling operation will occur at 8:10, wherein for the last 10 minutes, the CPU utilization is 80%. Then there will be a cool down period of 5 minutes. At 8:15, the CPU percentage is 80% so at this point another scaling operation will take place. Hence there will be a total of 3 instances running as part of the App Service Plan.`

Had to add additional answer as I could not comment more than 104 characters.