QuestionQ128
VerifiedPlanning and implementing a cloud solutionYou are running an application across multiple virtual machine instances within a managed instance group, with autoscaling enabled. The autoscaling policy adds instances to the group when CPU utilization exceeds 80%, continuing to add VMs until either the group reaches its maximum size of five VMs or CPU utilization drops back to 80%. The HTTP health check configured for the instances has an initial delay of 30 seconds.
You notice that the VM instances take approximately three minutes to become available to end users. When the instance group autoscales, it ends up provisioning more instances than are actually required to handle the current level of end-user traffic.
You want to ensure that the instance group size is properly maintained during autoscaling. What should you do?
Community Discussion