Examkin

Question

Difficulty: MediumPlanning Compute Engine Resources and Machine Types

A company is planning the Google Compute Engine architecture for two distinct workloads. Workload A is a nightly 4-hour batch processing job that is fault-tolerant and can resume from checkpoints if interrupted. Workload B is a 24/7 production relational database requiring steady, predictable vCPU performance with zero tolerance for abrupt termination. Which Compute Engine resource strategy minimizes total cost while meeting the operational requirements for both workloads?

  1. Provision Spot VMs for Workload A, and provision standard VM instances with Committed Use Discounts (CUDs) for Workload B.Answer
  2. B
    Provision Spot VMs for both Workload A and Workload B to maximize cost savings across all Compute Engine instances.
  3. C
    Provision standard VM instances with Sustained Use Discounts (SUDs) for Workload A, and provision Spot VMs with Committed Use Discounts (CUDs) for Workload B.
  4. D
    Migrate Workload A to Cloud Functions and deploy Workload B on Cloud Run with min-instances set to zero.

Answer

Provision Spot VMs for Workload A, and provision standard VM instances with Committed Use Discounts (CUDs) for Workload B.
Spot VMs are designed specifically for fault-tolerant, batch, or checkpointed workloads like Workload A, offering heavy discounts without risking service SLA. For continuous 24/7 workloads like Workload B, standard VM instances paired with Committed Use Discounts (CUDs) deliver predictable cost savings without preemption risk.

Step-by-Step Solution

1
Analyze the operational requirements of Workload A
Workload A runs for 4 hours nightly, is fault-tolerant, and supports checkpointing.
Fault-tolerant batch jobs that can handle interruptions are ideal candidates for Spot VMs, which provide discount rates up to 60-90% off standard pricing.
2
Analyze the operational requirements of Workload B
Workload B is a 24/7 production relational database requiring steady vCPU availability.
Continuous baseline stateful workloads must avoid preemption and can achieve optimal financial savings by committing to 1-year or 3-year Committed Use Discounts (CUDs).
3
Select the optimal combination
Pair Spot VMs for Workload A with CUD-backed standard VMs for Workload B.
This strategy maximizes cost optimization while strictly respecting the availability constraints of both services.

Key Concept

Compute Engine Resource Planning & Spot VM Suitability vs Committed Use Discounts
Rate this question

Topics

Planning Compute Engine Resources and Machine Types

More questions from this topic