Examkin

Question

Difficulty: Very hardDesigning Infrastructure for Business Requirements and Cost Optimization

A digital publishing company hosts a containerized REST API that serves breaking news content. Traffic is highly bursty and unpredictable: the service receives under 5 requests per minute during quiet periods, but surges to tens of thousands of requests per second during major breaking news events lasting 1 to 2 hours. The application reads from a regional relational database with standard transaction complexity. The company's chief financial officer requires an architectural design that eliminates baseline idle infrastructure costs while minimizing operational overhead. Which architecture best satisfies these business and cost requirements?

  1. Deploy the containerized application to Cloud Run with minimum instances set to zero, backed by a regional Cloud SQL instance with storage auto-scaling enabled.Answer
  2. B
    Deploy the containerized application to a Google Kubernetes Engine (GKE) Autopilot cluster and configure a Horizontal Pod Autoscaler (HPA) to scale down to 1 pod during quiet periods.
  3. C
    Deploy the containerized application to Cloud Run and replace the relational database with a multi-region Cloud Spanner instance provisioned with fixed processing units.
  4. D
    Deploy the application to a Compute Engine Managed Instance Group (MIG) and purchase 3-year Committed Use Discounts (CUDs) sized for the peak traffic footprint.

Answer

Deploying the containerized application to Cloud Run with minimum instances set to zero, backed by a regional Cloud SQL instance with storage auto-scaling enabled.
Cloud Run is a fully managed serverless platform that automatically scales containerized stateless HTTP applications down to zero instances when no traffic is present, eliminating compute costs during idle periods while providing instant scaling during breaking news spikes. Paired with Cloud SQL for regional relational storage, this provides a highly cost-effective design with minimal administrative overhead.

Step-by-Step Solution

1
Analyze compute workload characteristics and cost constraints
Workload is containerized, stateless, highly bursty, and idle most of the time.
Cloud Run allows scaling down to zero instances when idle, completely eliminating compute cost during quiet periods.
2
Evaluate database requirements against cloud database offerings
The application requires standard regional relational capabilities without multi-region global scale.
Cloud SQL provides cost-efficient relational database operations without the high fixed cost of Cloud Spanner.
3
Synthesize compute and database choices into a cost-optimized architecture
Combining Cloud Run (scale-to-zero) with regional Cloud SQL delivers minimal operational overhead and zero idle compute charges.
Meets both business goals of low cost during idle windows and minimal operational management.

Key Concept

Selecting serverless compute (Cloud Run scale-to-zero) and right-sized relational storage (Cloud SQL) to eliminate idle infrastructure costs for bursty workloads.
Estimated Time:2m 0s
Rate this question

Topics

Designing Infrastructure for Business Requirements and Cost Optimization