Examkin

Question

Difficulty: MediumConfigure Azure Monitor Autoscale Rules and Metrics

A backend data ingestion service runs on an Azure Virtual Machine Scale Set (VMSS) with a minimum capacity of 22 instances and a maximum capacity of 1010 instances. You need to configure autoscale rules based on the CPU percentage metric. The scale-out rule increases the capacity by 11 instance, and the scale-in rule decreases the capacity by 11 instance. Which of the following autoscale rule pairs will successfully prevent flapping when the VMSS is operating at its minimum capacity? (Select TWO)

  1. Scale-out when the average CPU percentage is greater than 90%90\%; scale-in when the average CPU percentage is less than 50%50\%.Answer
  2. B
    Scale-out when the average CPU percentage is greater than 80%80\%; scale-in when the average CPU percentage is less than 55%55\%.
  3. Scale-out when the average CPU percentage is greater than 75%75\%; scale-in when the average CPU percentage is less than 45%45\%.Answer
  4. D
    Scale-out when the average CPU percentage is greater than 70%70\%; scale-in when the average CPU percentage is less than 40%40\%, configured on a Free tier Azure App Service plan.

Answer

The correct autoscale rules are: (1) scale-out when average CPU percentage is greater than 90%90\% and scale-in when it is less than 50%50\%, and (2) scale-out when average CPU percentage is greater than 75%75\% and scale-in when it is less than 45%45\%.
The correct rule pairs prevent flapping because their scale-in thresholds are set lower than the post-scale-out average CPU load. Specifically, when starting from the minimum of 22 instances, scaling out to 33 instances reduces the average CPU to 23\frac{2}{3} of its pre-scale-out value. For a scale-out threshold of 90%90\%, the new average is 60%60\%, which is higher than the 50%50\% scale-in threshold. For a scale-out threshold of 75%75\%, the new average is 50%50\%, which is higher than the 45%45\% scale-in threshold.

Step-by-Step Solution

1
Formulate the condition to prevent flapping when capacity scales out from 22 instances to 33 instances.
The average CPU after scale-out must be higher than the scale-in threshold: Scale-in threshold<23×Scale-out threshold\text{Scale-in threshold} < \frac{2}{3} \times \text{Scale-out threshold}.
This ensures that adding an instance does not immediately lower the average CPU below the scale-in threshold.
2
Evaluate the metric thresholds for each candidate option under this condition.
For the rule with thresholds 90%90\% and 50%50\%, 23×90%=60%>50%\frac{2}{3} \times 90\% = 60\% > 50\% (safe). For the rule with thresholds 80%80\% and 55%55\%, 23×80%≈53.33%<55%\frac{2}{3} \times 80\% \approx 53.33\% < 55\% (flapping occurs). For the rule with thresholds 75%75\% and 45%45\%, 23×75%=50%>45%\frac{2}{3} \times 75\% = 50\% > 45\% (safe). For the rule with thresholds 70%70\% and 40%40\%, 23×70%≈46.67%>40%\frac{2}{3} \times 70\% \approx 46.67\% > 40\% (mathematically safe).
To identify which threshold combinations prevent flapping.
3
Check the resource hosting plan constraints.
Autoscale features require Standard, Premium, or Isolated App Service plans, and are not supported on the Free tier.
To eliminate configurations that cannot be implemented due to hosting plan limitations.

Key Concept

Azure Monitor Autoscale Flapping Prevention and Plan Requirements
Rate this question

Topics

Configure Azure Monitor Autoscale Rules and Metrics

More questions from this topic