Examkin

Question

Difficulty: MediumSelecting and Designing Compute and Application Platforms

An enterprise digital publishing company is modernizing its legacy content delivery architecture on Google Cloud. The solution requires hosting two distinct workloads:
1. A stateless REST API microservice that receives variable public HTTP traffic and must scale to zero during off-peak hours while minimizing operational infrastructure management.
2. A batch processing engine for raw media encoding that requires low-level Linux kernel sysctl parameters and custom OS kernel modules.

Which TWO compute platform choices should the Cloud Architect recommend to fulfill these requirements while optimizing operational efficiency and adherence to Google Cloud best practices? (Select TWO.)

  1. Deploy the stateless REST API microservice to Cloud Run to achieve containerized execution with scale-to-zero capability and zero server management overhead.Answer
  2. Deploy the batch processing engine to Compute Engine Virtual Machines using custom images configured with the required OS kernel modules and sysctl settings.Answer
  3. C
    Deploy the stateless REST API microservice to a multi-zone standard Google Kubernetes Engine (GKE) cluster to handle autoscaling and HTTP traffic routing.
  4. D
    Deploy the batch processing engine to Cloud Run by mounting the host kernel driver directory into the container environment.
  5. E
    Deploy both the REST API and the batch processing engine to App Engine Standard environment to unify serverless deployment.

Answer

The Cloud Architect should recommend deploying the stateless REST API microservice to Cloud Run and deploying the batch processing engine to Compute Engine Virtual Machines.
Selecting Cloud Run for the stateless REST API microservice satisfies the requirement for automatic scale-to-zero capability with minimal operational management. Selecting Compute Engine Virtual Machines for the media processing engine provides full administrative access to install custom kernel modules and configure sysctl parameters that managed serverless runtimes do not support.

Step-by-Step Solution

1
Analyze the requirements of the stateless REST API microservice
Identified requirements: stateless HTTP traffic, variable load, scaling to zero, minimal operational overhead.
Cloud Run is Google Cloud's fully managed serverless container platform designed specifically for stateless HTTP workloads, satisfying all operational and scaling requirements.
2
Analyze the requirements of the batch media encoding engine
Identified requirements: low-level Linux kernel sysctl modifications and custom kernel modules.
Managed serverless container runtimes (such as Cloud Run or App Engine Standard) run sandboxed environments and prohibit host kernel modifications. Compute Engine provides complete OS administrative access required for custom kernel modules.
3
Evaluate alternative options for compute misallocations
GKE introduces cluster management overhead for simple stateless APIs, while Cloud Run/App Engine Standard cannot support custom host OS kernel modifications.
Matching each workload to its ideal GCP compute platform balances operational simplicity with OS-level flexibility.

Key Concept

Compute Platform Selection Strategy (Cloud Run vs. Compute Engine vs. GKE)
Rate this question

Topics

Selecting and Designing Compute and Application Platforms