How Do You Control Compute Instances for a Service-Mesh Frontend?
Short answer
D Use a GKE Standard cluster with a fixed-size node pool and the Cloud Service Mesh add-on to cap compute instances while running the containerized services.
Question
You are deploying a new frontend service for an online game. The service was built using a micro-frontend architecture and consists of multiple containers that interact using a service mesh. You need to control the number of compute instances that are running at a given time. What should you do?
Key Takeaways
The exam tests the difference between controlling Pods or Cloud Run instances and controlling actual Compute Engine nodes: a GKE Standard fixed-size node pool sets the compute-instance count, whereas Autopilot and HPA/VPA manage workloads rather than underlying nodes.
This ACE question tests how to bound the number of compute instances for a containerized micro-frontend that relies on a service mesh. It establishes that a GKE Standard node pool with a fixed node count plus the Cloud Service Mesh add-on meets both requirements, while Autopilot and Cloud Run do not give the needed node-level control.
Common mistake: Learners often pick GKE Autopilot with vertical and horizontal Pod autoscaling (C) because it sounds Kubernetes-native, but Autopilot manages and scales nodes for you, so it does not let you pin the number of compute instances.
Expert Analysis
Why the Answer Is Correct
Option D uses a GKE Standard cluster, where you explicitly own and configure node pools. A node pool with a fixed number of nodes directly caps the number of Compute Engine instances running the workload. Enabling the Cloud Service Mesh add-on supplies the managed service mesh needed by the multiple containers, and GKE provides the orchestration layer for deploying those containers. This combination satisfies both stated requirements without hand-building VM-level container and mesh management.Why the Other Options Are Wrong
Option A does fix the number of instances in a managed instance group, but it relies on container-optimized VMs and a startup script that manually installs Istio, which is operationally heavy and not the managed Cloud Service Mesh path the question points toward. Option B uses Cloud Run min and max instances, but Cloud Run abstracts away Compute Engine instances and does not let you control node count; it is also not appropriate to deploy all micro-frontend containers into a single Cloud Run service. Option C uses GKE Autopilot, which automatically provisions and scales nodes, and HPA/VPA adjust Pods or resource requests rather than fixing the number of compute instances. Option D is therefore the only choice that cleanly combines managed service mesh with explicit node-level instance control.Community Comment Notes
Mr_MIXER007 connected the phrase "multiple containers that interact using a service mesh" to Cloud Service Mesh and "You need to control the number of compute instances that are running at a given time" to GKE Standard. That mapping matches the exam doctrine: GKE Standard is where you can pin the node pool size, while Autopilot removes that control. The vote data is unavailable, so the comment is best treated as a reasoning check rather than a majority signal.Exam Strategy
When a question asks you to control the number of compute instances, separate node-level control from workload-level autoscaling. If the scenario also requires a managed service mesh for many containers, prefer GKE Standard with the Cloud Service Mesh add-on over Cloud Run, Autopilot, or self-managed Istio on VMs.Official Reference
Exam Strategy
Look for the phrase fixed number of compute instances and map it to a fixed-size GKE Standard node pool, not to HPA/VPA or Autopilot. Then confirm the service-mesh requirement is met by the Cloud Service Mesh add-on rather than a manual Istio startup script.
Frequently Asked Questions
Why doesn't GKE Autopilot satisfy the fixed compute-instance requirement?
Autopilot provisions and scales nodes automatically, so you cannot pin a specific number of Compute Engine instances. HPA and VPA only adjust Pods or resource requests, not the node count.
Can Cloud Run minimum and maximum instances control Compute Engine nodes?
No. Cloud Run instance settings bound the number of serving instances, but Cloud Run abstracts the underlying infrastructure, so you do not control Compute Engine node counts.
Discussion 1 contributions
Compiled from public community threads and reader submissions, lightly edited for clarity. Handles and vote tallies are shown as recorded.
From public community threads 1
-
Mr_MIXER007 👍 1 Answered DD (Create a Google Kubernetes Engine (GKE) Standard cluster with a node pool that has a fixed number of nodes. Enable the Cloud Service Mesh add-on, and deploy the service.) «multiple containers that interact using a service mesh» = Cloud Service Mesh «You need to control the number of compute instances that are running at a given time» = GKE Standard.
Reader comments & corrections
No reader corrections yet — spotted an error or have a note? Share it below.
Related Analysis
Practice All ACE Questions
Access 84 questions with complete answers and detailed explanations.
View Full ACE Practice Test →