Question

Difficulty: MediumCreating Conceptual, Logical, and Physical Architectures

An enterprise insurance provider is translating its conceptual multi-tier architecture for a regional claims processing platform into a physical architecture on Google Cloud. The conceptual design specifies three core requirements: stateless web services that auto-scale to zero, a managed regional relational database supporting standard SQL transactions without global scaling needs, and a security perimeter preventing data exfiltration to unauthorized Google Cloud resources outside the organization. Which physical architecture on Google Cloud best fulfills these conceptual requirements while minimizing operational overhead and cost?

  1. Deploy Cloud Run for stateless web services, Cloud SQL for regional relational storage, and configure VPC Service Controls around the resources to prevent unauthorized data exfiltration.Answer
  2. B
    Deploy Cloud Run for stateless web services, Cloud Spanner for regional relational storage, and configure VPC Service Controls around the resources to prevent unauthorized data exfiltration.
  3. C
    Deploy Google Kubernetes Engine (GKE) Autopilot for stateless web services, Cloud SQL for regional relational storage, and configure VPC Service Controls around the resources to prevent unauthorized data exfiltration.
  4. D
    Deploy Cloud Run for stateless web services, Cloud SQL for regional relational storage, and grant standard IAM fine-grained roles to service accounts without configuring VPC Service Controls.

Answer

The correct architecture deploys Cloud Run for stateless microservices, Cloud SQL for regional relational data, and VPC Service Controls to prevent data exfiltration.
The solution properly translates conceptual architectural tiering to physical GCP components: Cloud Run fits the stateless scale-to-zero HTTP web tier, Cloud SQL satisfies regional relational transaction requirements cost-effectively, and VPC Service Controls fulfills security isolation against exfiltration.

Step-by-Step Solution

1
Map the compute requirement (stateless web services with scale-to-zero capability) to physical GCP services.
Cloud Run is selected over GKE because it provides serverless execution with auto-scaling to zero and minimum management overhead.
GKE introduces baseline node management costs and setup overhead for stateless services.
2
Map the storage requirement (regional relational database with SQL support) to physical GCP services.
Cloud SQL is selected over Cloud Spanner.
Cloud Spanner is designed for globally distributed database workloads, making it overly expensive and complex for single-region relational needs.
3
Map the network security requirement (preventing data exfiltration to outside resources).
VPC Service Controls is selected to establish a security perimeter.
IAM permissions alone cannot prevent data exfiltration by authorized users or compromised service accounts transferring data outside the boundary.

Key Concept

Translating Conceptual Requirements into Optimal Physical GCP Architecture
Rate this question