Examkin

Question

Difficulty: EasyConfigure Azure Monitor Autoscale Rules and Metrics

A queue-based backend worker application runs on an Azure App Service plan that is configured with a minimum of 55 instances. You set an autoscale rule to scale out by adding 11 instance when the average CPU percentage exceeds 80%80\%. You need to create a scale-in rule that removes 11 instance. To prevent autoscale flapping, what is the maximum CPU percentage threshold that should be set for the scale-in rule?

  1. A
    70%70\%
  2. B
    75%75\%
  3. 60%60\%Answer
  4. D
    80%80\%

Answer

A scale-in threshold of 60%60\% is the maximum value among the choices that is strictly less than the post-scale-out average CPU utilization of approximately 66.7%66.7\%, thereby preventing autoscale flapping.
The correct answer is 60%60\% because at the scale-out threshold of 80%80\% with 55 instances, the total CPU workload is 400%400\%. Adding a 6th6\text{th} instance distributes this workload to approximately 66.7%66.7\% CPU usage per instance. To prevent the scale-in rule from triggering immediately after scale-out (flapping), the scale-in threshold must be set below 66.7%66.7\%. The only option meeting this requirement is 60%60\%.

Step-by-Step Solution

1
Calculate the total CPU load at the scale-out threshold.
Total CPU load is 5 instances×80%=400%5 \text{ instances} \times 80\% = 400\%.
This represents the aggregate workload capacity required by the application when the scale-out rule is triggered.
2
Calculate the expected CPU utilization per instance after the scale-out event occurs.
New CPU utilization per instance is 400%/6 instances≈66.67%400\% / 6 \text{ instances} \approx 66.67\%.
Scaling out adds 11 instance, distributing the same workload across 66 instances.
3
Identify the scale-in threshold that is lower than the new per-instance CPU utilization.
The threshold must be strictly less than 66.67%66.67\%.
If the scale-in threshold is higher than or equal to 66.67%66.67\%, the scale-in rule will immediately trigger after the scale-out event, causing flapping. Among the options, 60%60\% is the only value below this limit.

Key Concept

Autoscale flapping prevention by calculating post-scaling metric distribution
Rate this question

Topics

Configure Azure Monitor Autoscale Rules and Metrics

More questions from this topic