Examkin

Soru

Zorluk: OrtaPlanning and Assessing Cloud Compute Solutions

An organization is designing a new cloud architecture for two distinct workloads:
1. A stateless containerized HTTP API that experiences unpredictable traffic spikes and requires automatic scaling down to zero instances during idle periods to minimize costs with zero node-management overhead.
2. A fault-tolerant nightly batch processing worker that processes queued images for several hours and can tolerate sudden node interruptions without data loss.

Which TWO Google Cloud compute deployment options should the cloud engineer select to meet these requirements while optimizing cost and operational efficiency? (Select TWO.)

  1. Deploy the stateless containerized HTTP API to Cloud Run.Cevap
  2. Deploy the fault-tolerant nightly batch processing workload on Spot Compute Engine virtual machines.Cevap
  3. C
    Deploy the stateless containerized HTTP API to Cloud Functions by refactoring the container into individual single-purpose event scripts.
  4. D
    Deploy the stateless containerized HTTP API to a GKE Standard cluster with manually configured fixed node pools to eliminate management overhead.
  5. E
    Deploy a non-fault-tolerant primary database on Spot Compute Engine virtual machines to reduce overall compute costs.

Cevap

The correct compute choices are deploying the stateless containerized HTTP API on Cloud Run and deploying the fault-tolerant nightly batch processing workload on Spot Compute Engine virtual machines.
Deploying the stateless containerized HTTP API on Cloud Run satisfies the requirement for container execution with scale-to-zero capabilities and minimal operational overhead. Deploying the fault-tolerant batch processing workload on Spot Compute Engine VMs leverages deeply discounted compute capacity suited for interruptible jobs.

Adım Adım Çözüm

1
Assess the requirements for Workload 1 (Stateless HTTP API).
The service is packaged as a container, requires scaling to zero, and demands zero infrastructure/node management overhead.
Cloud Run natively executes containerized web applications, automatically handles scaling from zero to match request volume, and abstracts away node management.
2
Assess the requirements for Workload 2 (Nightly Batch Worker).
The job runs for several hours, is fault-tolerant to unexpected interruptions, and prioritizes cost optimization.
Spot VMs provide deep cost savings for compute workloads that can accommodate preemptions.
3
Evaluate and eliminate suboptimal architectural options.
Refactoring containers for Cloud Functions, managing GKE Standard node pools manually, or hosting stateful non-fault-tolerant databases on Spot VMs violate operational or architectural constraints.
Selecting appropriate compute engines requires matching workload statefulness, fault tolerance, and management preferences to GCP service capabilities.

Anahtar Kavram

Selecting GCP Compute Services Based on Workload Characteristics (Cloud Run vs Spot VMs vs GKE)
Bu soruyu puanla

Konular

Planning and Assessing Cloud Compute Solutions

Bu konudan daha fazla soru