Examkin

Soru

Zorluk: Çok zorSelecting and Designing Compute and Application Platforms

An enterprise organization is architecting a public-facing API backend on Google Cloud to handle unpredictable burst traffic for a suite of stateless HTTP microservices. The business strategy mandates strict FinOps governance: infrastructure costs must scale strictly with incoming HTTP request volumes, scaling down to zero baseline cost during off-peak hours. Additionally, the organization has a lean Site Reliability Engineering (SRE) team that requires a fully managed environment to avoid node provisioning, OS patching, and Kubernetes cluster maintenance overhead. The services are packaged as standard OCI container images. Which compute platform architecture should a Cloud Architect recommend to satisfy these requirements?

  1. Deploy the containerized microservices directly to fully managed Cloud Run.Cevap
  2. B
    Provision a Google Kubernetes Engine (GKE) Autopilot cluster to deploy the microservice containers.
  3. C
    Deploy the containers to a Compute Engine regional Managed Instance Group (MIG) configured with CPU utilization autoscaling.
  4. D
    Deploy the microservices on Compute Engine Virtual Machines backed by 3-year Committed Use Discounts (CUDs) to baseline capacity.

Cevap

Deploying the containerized microservices directly to fully managed Cloud Run is the optimal architecture.
Deploying the microservices to fully managed Cloud Run fulfills all architectural constraints. Cloud Run provides a serverless platform that natively runs OCI container images, automatically scales up or down to zero instances based on incoming request traffic, eliminates baseline operational costs during off-peak idle hours, and completely removes cluster and OS maintenance burdens from the SRE team.

Adım Adım Çözüm

1
Analyze technical and operational workload requirements
Workload requires stateless HTTP container execution, fast auto-scaling with zero baseline cost, and minimal operational maintenance.
The SRE team is small and the business mandates scaling costs strictly with request volume without fixed baseline charges.
2
Evaluate GCP compute platform candidates against scaling and management requirements
Compute Engine MIGs and GKE clusters incur baseline instance/cluster management overhead and continuous baseline infrastructure costs.
MIGs require OS management and cannot scale to zero; GKE introduces cluster management overhead and baseline cluster fees.
3
Select optimal serverless container platform
Cloud Run runs standard OCI containers natively, provides pay-per-request pricing, scales seamlessly from zero to high concurrency, and handles all underlying infrastructure management automatically.
Cloud Run fulfills every constraint: stateless container execution, zero cost during idle periods, and zero cluster node administration.

Anahtar Kavram

Selecting Serverless Container Compute (Cloud Run) vs. Managed Kubernetes (GKE) for Stateless Microservices
Tahmini Süre:2m 0s
Bu soruyu puanla

Konular

Selecting and Designing Compute and Application Platforms