Examkin

Soru

Zorluk: OrtaSelecting and Designing Compute and Application Platforms

An enterprise HR platform is migrating its background check system to Google Cloud. The system consists of two distinct components:

1. Candidate Web Portal: A stateless HTTP API receiving candidate form submissions. Traffic is highly unpredictable, spiking during business hours and dropping to near zero overnight. The platform team requires zero infrastructure management overhead and automatic scaling down to zero.
2. Verification Rendering Engine: A continuous batch process that performs document validation and rendering. It requires custom C++ Linux socket tuning at the operating system level and runs continuously 24/7 with steady compute demands.

Which TWO architectural decisions should the Cloud Architect recommend to optimize operational efficiency and cost?

  1. Deploy the Candidate Web Portal on Cloud Run.Cevap
  2. Deploy the Verification Rendering Engine on Compute Engine Virtual Machines backed by Committed Use Discounts.Cevap
  3. C
    Provision a multi-zone Google Kubernetes Engine (GKE) Standard cluster to host the Candidate Web Portal.
  4. D
    Host the Verification Rendering Engine on Cloud Run using maximum concurrency limits.
  5. E
    Purchase 3-year Committed Use Discounts to cover the peak capacity of the Candidate Web Portal.

Cevap

Deploy the Candidate Web Portal on Cloud Run, and deploy the Verification Rendering Engine on Compute Engine Virtual Machines backed by Committed Use Discounts.
Cloud Run provides a fully managed serverless execution environment ideal for bursty, stateless HTTP applications requiring scale-to-zero capability without infrastructure management. Compute Engine virtual machines provide full control over the operating system environment needed for custom kernel tuning, while Committed Use Discounts offer maximum cost savings for sustained 24/7 workloads.

Adım Adım Çözüm

1
Analyze the requirements for Component 1 (Candidate Web Portal)
The workload is stateless, HTTP-based, bursty, requires zero infrastructure management, and needs scale-to-zero capabilities.
Cloud Run is the optimal Google Cloud compute platform for stateless containerized web services requiring scale-to-zero and fully managed operations.
2
Analyze the requirements for Component 2 (Verification Rendering Engine)
The workload requires custom Linux OS kernel socket modifications and runs continuously 24/7 with steady demand.
Compute Engine provides complete OS control for kernel-level customizations, while Committed Use Discounts significantly lower costs for steady 24/7 usage.
3
Evaluate and reject misallocated alternatives
GKE adds unnecessary management overhead for simple microservices; Cloud Run does not allow low-level kernel modifications; Committed Use Discounts on bursty scale-to-zero workloads lead to financial waste.
Matching compute options directly to workload statefulness, customization limits, and scaling characteristics ensures architectural alignment with GCP best practices.

Anahtar Kavram

Compute platform selection based on scaling behavior, OS-level customization, operational overhead, and cost optimization models.
Bu soruyu puanla

Konular

Selecting and Designing Compute and Application Platforms