Cloud Digital Leader Why Cloud Technology Can Transform Business Practice Question

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  Cluster Autoscaler

  Why it's wrong here

  Cluster Autoscaler is a GKE component that adjusts the number of nodes in the cluster by adding or removing machines based on pending pods that cannot be scheduled due to insufficient capacity. It does not modify the replica count of a Deployment or respond to application-level metrics such as request latency or queue depth. For scaling pods on custom metrics, you need the Horizontal Pod Autoscaler (HPA) backed by a metrics adapter, not the Cluster Autoscaler.

• ✓

  Custom Metrics Stackdriver Adapter

  Why this is correct

  The Custom Metrics Stackdriver Adapter is the correct component because it implements the Kubernetes custom metrics API and exposes Stackdriver (now Cloud Monitoring) metrics to the Horizontal Pod Autoscaler. Without this adapter, HPA can only see built-in CPU and memory metrics, not application-specific metrics like orders per second or checkout latency. Installing and configuring this adapter allows HPA to query custom metrics and scale pod replicas accordingly, making it essential for autoscaling on arbitrary business metrics.

• ✗

  Vertical Pod Autoscaler (VPA)

  Why it's wrong here

  Vertical Pod Autoscaler (VPA) is wrong because it changes the CPU and memory requests and limits of individual pods to better fit their usage, but it never changes the number of pod replicas. VPA is a scaling mechanism for resource efficiency, not for handling variable traffic by adding more instances. Even if it could read custom metrics, it would adjust resource settings, not scale out horizontally, so it cannot replace HPA for custom-metric-based autoscaling.

• ✗

  Cloud Load Balancing

  Why it's wrong here

  Cloud Load Balancing simply distributes incoming network traffic across backend pods or instances based on routing rules and health checks. It does not inspect application metrics, make scaling decisions, or trigger changes to the number of replicas. A load balancer is a networking layer component that directs traffic to whatever endpoints exist at the moment; it is not a controller that reacts to demand by creating or deleting pods.

• ✓

  Horizontal Pod Autoscaler (HPA)

  Why this is correct

  Horizontal Pod Autoscaler (HPA) is a correct and necessary component because it is the native Kubernetes controller that automatically adjusts the number of pod replicas in a Deployment or StatefulSet based on observed metrics. However, by default HPA only supports resource metrics such as CPU and memory from the Metrics Server. To use custom business metrics from Stackdriver, HPA must be paired with the Custom Metrics Stackdriver Adapter, which feeds those metrics into the API that HPA consumes—so HPA alone is not the complete solution, but it is the right scaling mechanism.

Go deeper

Related to this question