Examkin

Question

Difficulty: Very hardScale Azure App Service Web Apps

A logistics tracking solution uses an Azure App Service Web App to process real-time GPS telemetry. The web app is currently hosted on the Free (F1) pricing tier. During morning dispatch hours, telemetry processing causes the memory usage to spike up to 85%, resulting in out-of-memory errors. You must configure the App Service plan and autoscale rules to dynamically scale out the web app when the Memory Percentage metric exceeds 80% for 10 minutes, while minimizing costs and preventing autoscale flapping.

Which two actions should you perform? (Select TWO)

  1. Scale up the App Service plan to the Standard (S1) pricing tier.Answer
  2. Configure the scale-in rule to trigger when the Memory Percentage is less than 35%.Answer
  3. C
    Scale up the App Service plan to the Basic (B1) pricing tier.
  4. D
    Configure the scale-in rule to trigger when the Memory Percentage is less than 65%.

Answer

To configure the scaling correctly, you must scale up the App Service plan to the Standard (S1) pricing tier and configure the scale-in rule to trigger when the Memory Percentage is less than 35%.
Scaling up to the Standard (S1) tier is required because it is the lowest-cost tier that supports autoscale rules. Setting the scale-in threshold to 35% prevents flapping because scaling out from 1 to 2 instances at 80% load reduces the average load per instance to 40%. A scale-in threshold of 35% is lower than this post-scale-out value, ensuring the system remains scaled out until the overall demand decreases.

Step-by-Step Solution

1
Determine the minimum App Service pricing tier that supports metric-based autoscale rules.
The Standard (S1) tier is identified as the most cost-effective tier supporting autoscale. The Free (F1) tier has no scaling, and the Basic (B1) tier only supports manual scaling.
This establishes the target pricing tier required to implement dynamic scaling rules.
2
Calculate the post-scale-out metric value to determine the flapping boundary.
Scaling out from 1 instance to 2 instances at 80% memory usage distributes the load, dropping the average memory percentage to approximately 40% (80%÷2=40%80\% \div 2 = 40\%) across the instances.
This calculation is necessary to determine the threshold below which the scale-in rule must be set to prevent immediate scale-back.
3
Evaluate the scale-in thresholds against the calculated post-scale-out load.
A threshold of 65% is higher than the post-scale-out load of 40%, which would trigger an immediate scale-in (flapping). A threshold of 35% is safely below 40%, preventing the scale-in until load drops further.
Selecting the correct threshold ensures stable scaling behavior under peak loads.

Key Concept

Configuring autoscale rules and selecting the appropriate pricing tier for Azure App Service to prevent flapping.
Rate this question

Topics

Scale Azure App Service Web Apps

More questions from this topic