Examkin

Soru

Zorluk: OrtaPlanning Compute Engine Resources and Machine Types

A financial services company is planning to deploy a dedicated transaction reporting microservice on Google Cloud Compute Engine. The service runs continuously 24/7 and requires an uninterrupted baseline of 4 vCPUs and 26 GB of RAM to meet strict Service Level Agreements (SLAs). Standard machine types offer either insufficient memory or excess unneeded vCPUs for this specific memory ratio. Which compute provisioning strategy should you choose to meet the operational requirements while minimizing costs?

  1. Provision a custom machine type with 4 vCPUs and 26 GB of RAM on Compute Engine, and purchase a Committed Use Discount (CUD) for the baseline capacity.Cevap
  2. B
    Provision a Spot VM instance with 4 vCPUs and 26 GB of RAM to achieve maximum baseline cost reduction.
  3. C
    Deploy the application to Google Cloud Functions to automatically manage memory scaling and eliminate VM overhead.
  4. D
    Deploy the workload to a GKE Autopilot cluster configured with a standard preset N2 machine family node pool.

Cevap

Provision a custom machine type with 4 vCPUs and 26 GB of RAM on Compute Engine, and purchase a Committed Use Discount (CUD) for the baseline capacity.
Custom Machine Types enable exact tailoring of CPU and RAM allocations when standard predefined shapes do not match the required memory ratio. Purchasing a Committed Use Discount (CUD) provides deep pricing discounts for steady-state 24/7 baseline capacity without risking preemption.

Adım Adım Çözüm

1
Analyze workload resource requirements and operational constraints.
The workload runs 24/7 as a steady baseline, requires uninterrupted SLA execution, and needs a specific non-standard ratio of 4 vCPUs to 26 GB RAM.
Determining exact resource constraints prevents paying for excess vCPUs found in predefined machine shapes.
2
Select the appropriate Compute Engine machine configuration.
Create a Custom Machine Type with 4 vCPUs and 26 GB RAM.
Custom Machine Types enable tailor-made CPU and memory configurations when standard predefined shapes do not fit optimal resource ratios.
3
Select the cost optimization model for steady baseline usage.
Apply a Committed Use Discount (CUD) for the baseline compute resources.
Committed Use Discounts offer substantial savings for predictable 24/7 workloads without operational preemption risk.

Anahtar Kavram

Custom Machine Types and Committed Use Discounts (CUDs) for continuous Compute Engine workloads
Bu soruyu puanla

Konular

Planning Compute Engine Resources and Machine Types

Bu konudan daha fazla soru