Prepare and test your skills

Prepare and test your skills

Free sample question with answers and detailed explanation below. The Associate Cloud Engineer practice bank has an additional 2,014 practice questions.Try 10 free realistic questions: instant score, no credit card, and no account required.
An operations team is deploying a microservice workload to a Google Kubernetes Engine (GKE) cluster.
They want to utilize Vertical Pod Autoscaling (VPA) to analyze historical container resource usage and generate recommended CPU and memory requests. Because the service is sensitive to disruption, the team must ensure that running Pods are never evicted or restarted by the autoscaler while collecting these recommendations.
Which VPA updateMode should the team configure?
The Auto update mode instructs the Vertical Pod Autoscaler (VPA) to manage container resource sizing automatically throughout the entire workload lifecycle. It sets CPU and memory requests when new Pods are scheduled and actively evicts running Pods when their current allocations deviate significantly from calculated recommendations.
Auto mode is inappropriate.PodDisruptionBudget (PDB) to throttle evictions, but evictions still take place.The goal was to fully automate container right-sizing dynamically at runtime, where the application architecture is fully stateless, horizontally resilient, and tolerant of frequent rolling restarts.
In Google Kubernetes Engine (GKE), the Vertical Pod Autoscaler (VPA) with updateMode: "Off" enables resource analysis without actively applying changes to running or new Pods. In this mode, the VPA recommender component monitors historical and current CPU and memory consumption, calculates optimal request values, and writes these recommendations directly to the status subresource of the VerticalPodAutoscaler object.

Setting updateMode: "Off" is the standard, risk-free recommendation mechanism in GKE. It fulfills the goal of capturing sizing metrics from actual workloads without altering the Pod lifecycle.
The Initial update mode configures the Vertical Pod Autoscaler (VPA) to apply CPU and memory recommendations only during the Pod creation phase (via an admission webhook). Once a Pod is running, VPA will not evict it to update its resource requests.
Initial mode alters workload resources without manual verification.The team wanted newly scheduled Pods to immediately adopt VPA recommendations while ensuring that already running Pods are never evicted mid-execution.
The Recreate update mode causes the Vertical Pod Autoscaler (VPA) to assign resource requests upon Pod creation and evict running Pods whenever their resource consumption necessitates updated requests.
Auto, Recreate directly evicts active Pods whose container request sizes need adjustments so that the deployment controller will recreate them.Auto in current Kubernetes versions when in-place resource resizing is not enabled.The team required running Pods to be restarted whenever recommended resource allocations changed, ensuring that outdated container limits are promptly removed from the cluster.
Answer them yourself and get your score the moment you finish. No account needed.
Full access includes
Keep the momentum going with these hand-picked practice scenarios
Last updated