Soru

Zorluk: OrtaDesigning Infrastructure for Business Requirements and Cost Optimization

A global logistics platform is migrating its backend operational workloads to Google Cloud. The system consists of stateless HTTP REST microservices with highly variable, bursty traffic patterns, alongside a 300 GB300\text{ GB} relational database supporting standard regional ACID transactions. The company's primary business objective is to minimize baseline infrastructure costs and operational management overhead. Which TWO architecture decisions should you recommend to achieve these cost optimization goals? (Select TWO.)

  1. Deploy the stateless REST microservices to Cloud Run to eliminate baseline idle compute costs through automatic scale-to-zero capability.Cevap
  2. B
    Deploy the stateless REST microservices to a GKE Standard cluster with multi-zone node pools to ensure custom container runtime orchestration.
  3. Migrate the relational database to Cloud SQL for PostgreSQL to fulfill single-region ACID requirements while minimizing baseline licensing and management expense.Cevap
  4. D
    Migrate the relational database to Cloud Spanner to ensure enterprise-grade horizontal scalability and seamless multi-region expansion.

Cevap

Deploy the stateless microservices on Cloud Run and host the relational database using Cloud SQL for PostgreSQL.
To minimize baseline infrastructure costs for variable stateless microservices and a moderate-sized regional database, serverless compute and regional database offerings are optimal. Cloud Run provides scale-to-zero capabilities for HTTP services, eliminating idle infrastructure expense. Cloud SQL for PostgreSQL delivers fully managed ACID compliance tailored to regional database requirements at a fraction of the cost of globally distributed database services.

Adım Adım Çözüm

1
Analyze compute requirements and cost profile.
Stateless REST microservices with variable, bursty traffic benefit from serverless scale-to-zero compute (Cloud Run) rather than paying for idle compute nodes in a managed Kubernetes cluster (GKE).
Cloud Run eliminates baseline compute costs during inactive periods without management overhead.
2
Analyze storage and database requirements and cost profile.
A 300 GB300\text{ GB} single-region relational dataset requiring standard ACID compliance is best served by Cloud SQL for PostgreSQL.
Cloud Spanner is designed for global scale across multiple regions; using it for a single-region moderate workload introduces significant unnecessary minimum node costs.

Anahtar Kavram

Matching serverless compute and regional relational databases to workload scale and traffic patterns to optimize Google Cloud infrastructure costs.
Bu soruyu puanla