Examkin

Question

Difficulty: MediumPlanning Compute Engine Resources and Machine Types

A software enterprise is planning the infrastructure for a continuous, mission-critical API gateway on Google Cloud that requires non-standard resource allocation of 10 vCPUs and 40 GB of RAM, running 24 hours a day, 7 days a week. The workload cannot tolerate unexpected interruptions or instance terminations. The finance team mandates the most cost-effective architecture for a guaranteed 3-year commitment. Which deployment strategy should you recommend?

  1. Provision Compute Engine Virtual Machines using a custom machine type with 10 vCPUs and 40 GB of RAM, and purchase a 3-year Committed Use Discount for the required compute capacity.Answer
  2. B
    Provision Compute Engine Spot Virtual Machines configured with a custom machine type of 10 vCPUs and 40 GB of RAM to achieve the lowest hourly rate.
  3. C
    Deploy the application to Cloud Run with container limits set to 10 vCPUs and 40 GB RAM to minimize infrastructure management costs.
  4. D
    Deploy the workload to Google Kubernetes Engine Autopilot clusters so Google Cloud can dynamically provision compute resources and apply committed use pricing automatically.

Answer

Provision Compute Engine Virtual Machines using a custom machine type with 10 vCPUs and 40 GB of RAM, combined with a 3-year Committed Use Discount.
Configuring a custom machine type with exactly 10 vCPUs and 40 GB of RAM ensures the instance is tailored to the exact workload demand without over-provisioning unused resources. Purchasing a 3-year Committed Use Discount (CUD) maximizes savings for steady-state, 24/7 uninterruptible workloads.

Step-by-Step Solution

1
Analyze workload resource ratio and uninterrupted availability requirements.
The requirement calls for 10 vCPUs and 40 GB of RAM (1:4 ratio), running 24/7 without tolerable downtime.
Matching non-standard ratios avoids paying for wasted vCPUs or RAM present in rigid standard machine types, while uninterrupted requirements eliminate Spot/Preemptible VMs.
2
Evaluate pricing structure options for long-term continuous Compute Engine workloads.
A 3-year Committed Use Discount (CUD) offers the deepest discount for baseline Compute Engine vCPU and memory capacity.
Committed Use Discounts require a commitment of 1 or 3 years and are ideal for steady-state workloads with predictable resource needs.

Key Concept

Custom Machine Types and Committed Use Discounts (CUDs)
Estimated Time:1m 30s
Rate this question

Topics

Planning Compute Engine Resources and Machine Types

More questions from this topic