Examkin

Question

Difficulty: MediumPlanning Compute Engine Resources and Machine Types

A financial software company runs an internal batch calculations service on Google Compute Engine. The application requires exactly 6 vCPUs and 28 GB of memory to process daily data efficienty, but none of the predefined machine types match this resource ratio. The workload runs continuously 24 hours a day, 7 days a week, and cannot tolerate unexpected preemptions or terminations. Which Compute Engine machine configuration and pricing model should you select to minimize operational costs while satisfying all system requirements?

  1. Provision a Custom Machine Type with 6 vCPUs and 28 GB of memory, and purchase a Committed Use Discount (CUD) for the baseline usage.Answer
  2. B
    Provision a Custom Machine Type with 6 vCPUs and 28 GB of memory using Spot VMs to achieve maximum cost savings.
  3. C
    Provision a predefined n2-standard-8 instance type and rely on automatic Sustained Use Discounts (SUD) to optimize compute costs.
  4. D
    Deploy the application to Cloud Run with custom container CPU and memory allocations to eliminate virtual machine management.

Answer

Provision a Custom Machine Type with 6 vCPUs and 28 GB of memory, and purchase a Committed Use Discount (CUD) for the baseline usage.
Configuring a Custom Machine Type allows exact allocation of 6 vCPUs and 28 GB of RAM without paying for unused resources of larger predefined types. Pairing continuous, non-interruptible 24/7 workloads with Committed Use Discounts provides the highest cost savings while guaranteeing instance availability.

Step-by-Step Solution

1
Analyze resource ratio requirements
Identified that predefined machine families do not offer a 6 vCPU to 28 GB memory ratio.
Google Compute Engine allows creating Custom Machine Types to fit exact CPU and memory specifications, preventing over-provisioning.
2
Evaluate workload availability and cost optimization constraints
Workload runs continuous 24/7 steady-state and cannot tolerate interruption.
Committed Use Discounts (CUDs) provide deep discounts for steady-state workloads with 1-year or 3-year commitments without preemption risk.

Key Concept

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

Topics

Planning Compute Engine Resources and Machine Types

More questions from this topic