PROFESSIONAL-CLOUD-DEVELOPER · Question #15
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
The correct answer is B. Decrease the cool-down period for instances added to the group. D. Decrease the target CPU usage for the instance group autoscaler.. The root cause is that the autoscaler waits too long before scaling out, and then takes too long to stabilize after adding instances. Two changes fix this: (D) Decrease the target CPU usage - by setting a lower CPU threshold (e.g., 50% instead of 80%), the autoscaler triggers sca
Question
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.)
Options
- AAdd the label "AUTOSCALE" to the instance group template.
- BDecrease the cool-down period for instances added to the group.
- CIncrease the target CPU usage for the instance group autoscaler.
- DDecrease the target CPU usage for the instance group autoscaler.
- ERemove the health-check for individual VMs in the instance group.
How the community answered
(49 responses)- A2% (1)
- B80% (39)
- C12% (6)
- E6% (3)
Explanation
The root cause is that the autoscaler waits too long before scaling out, and then takes too long to stabilize after adding instances. Two changes fix this: (D) Decrease the target CPU usage - by setting a lower CPU threshold (e.g., 50% instead of 80%), the autoscaler triggers scale-out earlier, before CPU saturation causes latency spikes. (B) Decrease the cool-down period - the cool-down is the time the autoscaler waits after adding an instance before it evaluates metrics again; shortening it allows faster successive scaling decisions. Increasing the target CPU (C) would make latency worse. Options A and E are irrelevant to autoscaling behavior.
Topics
Community Discussion
No community discussion yet for this question.