Exam Professional Cloud Architect topic 1 question 140 discussion

Answer is D. num_undelivered_messages metric can indicate if subscribers are keeping up with message submissions. https://cloud.google.com/pubsub/docs/monitoring#monitoring_the_backlog

Direct answer - D https://cloud.google.com/kubernetes-engine/docs/samples/container-pubsub-horizontal-pod-autoscaler apiVersion: autoscaling/v2beta2 kind: HorizontalPodAutoscaler metadata: name: pubsub spec: minReplicas: 1 maxReplicas: 5 metrics: - external: metric: name: pubsub.googleapis.com|subscription|num_undelivered_messages selector: matchLabels: resource.labels.subscription_id: echo-read target: type: AverageValue averageValue: 2 type: External scaleTargetRef: apiVersion: apps/v1 kind: Deployment name: pubsub

Subscription Metric: Scaling based on the subscription/num_undelivered_messages metric directly ties the scaling behavior to the number of unprocessed messages in Pub/Sub. This ensures that your application scales out when there are more messages to process and scales in when the queue is short. Relevant Metric: This metric is relevant for an I/O-intensive application that processes messages from Pub/Sub, ensuring that the scaling is directly responsive to the message processing demand.

D is the correct answer: subscription/num_undelivered_messages directly indicates the number of messages waiting to be processed, making it a perfect indicator of the workload on the application.

https://cloud.google.com/kubernetes-engine/docs/tutorials/autoscaling-metrics#pubsub here . so this is D

Option C (--enable-autoscaling flag for the entire cluster): Enabling autoscaling at the cluster level doesn't provide fine-grained control over scaling individual pods or deployments based on specific workload metrics.

Can someone tell that these answers are right like people voted D but answer is C.

Answer D

B & C refer to using auto scaler with custom metrics but "Cluster autoscaler makes these scaling decisions based on the resource requests (rather than actual resource utilization) of Pods running on that node pool's nodes. " A makes no sense as it defines a min and max that we don't know D seems like part of the solution so D.

is D - https://cloud.google.com/kubernetes-engine/docs/tutorials/autoscaling-metrics#pubsub

I vot for B. Configure a Kubernetes autoscaling deployment based on the subscription/push_request_latencies metric. as the metric should be based on latency instead of num_undelivered_messages metric in D.

D is the correct answer

D is the correct one

D is ok

D https://cloud.google.com/kubernetes-engine/docs/tutorials/autoscaling-metrics#pubsub

This seems relevant https://cloud.google.com/kubernetes-engine/docs/tutorials/autoscaling-metrics#pubsub even if it uses Deployment + HorizontalPodAutoscaler which is not mentioned in the context of the question/answer

I think it wrong for application to be in pod it should be a deployment, and deployment would scale