Soru

Zorluk: OrtaScale Azure App Service Web Apps

An application named PromoCampaignPortal is currently hosted on a Basic (B1B1) App Service plan. During marketing campaigns, the application experiences significant memory spikes. You want to implement autoscaling to automatically handle this demand, while ensuring that the scaling behavior remains stable and does not cause rapid, repeated scaling actions (flapping).

Which two of the following actions must you perform to meet these requirements?

  1. Scale up the App Service plan to the Standard (S1S1) pricing tier.Cevap
  2. Configure a scale-out rule to add an instance when memory usage exceeds 80%80\% and a scale-in rule to remove an instance when memory usage falls below 40%40\%.Cevap
  3. C
    Configure autoscaling directly on the existing Basic (B1B1) App Service plan by setting the minimum instance count to 22.
  4. D
    Configure a scale-out rule to add an instance when memory usage exceeds 80%80\% and a scale-in rule to remove an instance when memory usage falls below 75%75\%.

Cevap

To configure autoscaling for the application, you must scale up the App Service plan to the Standard (S1S1) pricing tier, as the Basic (B1B1) tier does not support autoscale. In addition, you must set a scale-out rule to add an instance when memory usage exceeds 80%80\% and a scale-in rule to remove an instance when memory usage falls below 40%40\% to maintain a safe margin and prevent autoscale flapping.
To enable rules-based autoscaling, the App Service plan must be scaled up to at least the Standard (S1S1) pricing tier, as the Basic (B1B1) tier only supports manual scaling. In addition, when configuring scale-out and scale-in rules, there must be a sufficient margin between the thresholds to prevent flapping. Setting the scale-out rule to trigger when memory exceeds 80%80\% and the scale-in rule to trigger when memory drops below 40%40\% ensures that the reduction in average memory usage per instance after scaling out does not immediately trigger a scale-in action.

Adım Adım Çözüm

1
Evaluate the current App Service pricing tier for autoscale compatibility.
Identify that the Basic (B1B1) plan only supports manual scale-out (up to 33 instances) and does not support autoscale rules.
To use autoscale rules, the App Service plan must first be scaled up to at least the Standard (S1S1) tier.
2
Scale up the App Service plan.
The App Service plan is updated to the Standard (S1S1) pricing tier, enabling autoscale capabilities.
Enables the Azure Monitor autoscale engine to manage the instance count dynamically.
3
Determine the appropriate autoscale metric thresholds to avoid flapping.
Establish a scale-out threshold at 80%80\% memory usage and a scale-in threshold at 40%40\% memory usage.
A sufficient gap between the scale-out and scale-in thresholds prevents the system from immediately scaling back in once the load is distributed to the new instance, thereby preventing flapping.

Anahtar Kavram

Autoscaling configuration requirements and flapping prevention in Azure App Service
Bu soruyu puanla