Question

Difficulty: MediumPlanning Google Kubernetes Engine (GKE) Cluster Architectures

A renewable energy company is planning a Google Kubernetes Engine (GKE) cluster architecture to support two distinct cloud workloads:
1. A stateless telemetry ingestion API service with fluctuating traffic, where the operations team requires zero node management overhead and pod-level resource billing.
2. A fault-tolerant, stateless batch calculation engine that processes sensor data off-peak, where reducing compute cost is the primary constraint.

Which TWO cluster configuration decisions should the cloud engineering team implement? (Select TWO.)

  1. Provision a GKE Autopilot cluster for the telemetry ingestion API workload.Answer
  2. Create a Spot VM node pool for the off-peak batch calculation engine workload.Answer
  3. C
    Deploy a GKE Standard cluster using manually sized static node pools for the telemetry ingestion API workload.
  4. D
    Configure standard On-Demand Compute Engine node pools with persistent storage for the batch calculation engine workload to prevent preemption.
  5. E
    Configure the Horizontal Pod Autoscaler (HPA) to scale the node count of the cluster when batch processing queue depth increases.

Answer

The team should provision a GKE Autopilot cluster for the telemetry ingestion API service and create a Spot VM node pool for the off-peak batch calculation engine workload.
Provisioning a GKE Autopilot cluster satisfies the requirement for zero node management overhead and pod-based billing. Utilizing a Spot VM node pool for the stateless, fault-tolerant batch calculation engine drastically reduces operational compute costs.

Step-by-Step Solution

1
Analyze the operational requirements for the telemetry ingestion API workload.
GKE Autopilot is selected because it completely removes node management responsibilities from the operations team while enforcing pod-level billing.
GKE Autopilot automates node provisioning, updates, and maintenance.
2
Analyze the financial and operational constraints for the batch calculation workload.
A Spot VM node pool is selected for batch processing.
Spot VMs deliver substantial cost savings for stateless, fault-tolerant workloads that can tolerate unexpected instance preemptions.

Key Concept

Planning GKE cluster operational models (Autopilot vs. Standard) and node pool provisioning strategies (Spot VMs)
Rate this question