You configured your Compute Engine instance group to scale automatically according to overall CPU usage.
However, your application's response latency increases sharply before the cluster has finished adding up instances.
You want to provide a more consistent latency experience for your end users by changing the configuration of the instance group autoscaler.
Which two configuration changes should you make? (Choose two.)
Click on the arrows to vote for the correct answer
A. B. C. D. E.AC.
To provide a more consistent latency experience for end-users by changing the configuration of the instance group autoscaler, two configuration changes are required.
Decrease the cool-down period for instances added to the group: When a new instance is added to the group, it takes some time to configure and start up before it becomes fully operational. During this period, the instance may not be able to handle requests at full capacity, leading to a slower response time for end-users. The cool-down period is the time the autoscaler waits before adding another instance after the previous one. Decreasing the cool-down period will allow new instances to become operational more quickly, thereby reducing the latency.
Increase the target CPU usage for the instance group autoscaler: The target CPU usage is the average CPU utilization of all instances in the group that the autoscaler aims to maintain. When the CPU utilization exceeds the target, the autoscaler adds new instances to the group. Increasing the target CPU usage will cause the autoscaler to add instances sooner, before the CPU utilization reaches a level that could affect the latency.
Therefore, options B and C are the correct configuration changes to make.
Option A is incorrect because adding a label to the instance group template will not affect the autoscaler's behavior.
Option D is incorrect because decreasing the target CPU usage will cause the autoscaler to add instances later, which could worsen the latency.
Option E is incorrect because removing the health-check for individual VMs in the instance group may cause the autoscaler to keep instances running even if they are not healthy, which could affect the overall performance of the application.