Examkin

Question

Difficulty: HardPlanning Compute Engine Resources and Machine Types

An organization is planning its Google Compute Engine infrastructure for two new enterprise services with distinct operational profiles:

1. Workload 1: An offline rendering and data processing pipeline that runs daily. The application is stateless, fault-tolerant, and designed to checkpoint progress so it can be safely paused and resumed.
2. Workload 2: A core relational database instance requiring heavy RAM usage per vCPU that must run continuously 24/7 with zero unannounced downtime.

Which TWO architectural decisions correctly align with Google Cloud resource planning and pricing best practices for these workloads? (Select TWO.)

  1. Provision Spot VMs for Workload 1 to minimize compute overhead for fault-tolerant batch jobs.Answer
  2. Provision a memory-optimized machine family instance combined with a 1-year or 3-year Committed Use Discount (CUD) for Workload 2.Answer
  3. C
    Provision Spot VMs for Workload 2 to eliminate baseline instance infrastructure costs for persistent database hosting.
  4. D
    Rely primarily on default Sustained Use Discounts (SUDs) rather than Committed Use Discounts to maximize savings on predictable 3-year production database workloads.

Answer

The two correct architectural decisions are: using Spot VMs for the stateless, fault-tolerant offline data processing pipeline (Workload 1), and provisioning a memory-optimized machine family with Committed Use Discounts for the continuous 24/7 relational database (Workload 2).
Deploying Spot VMs for fault-tolerant, checkpoint-enabled batch jobs takes advantage of extreme cost discounts without risking job completion. For steady-state 24/7 databases needing high RAM, selecting a memory-optimized machine family with Committed Use Discounts maximizes performance while securing the deepest contractual cost reduction.

Step-by-Step Solution

1
Analyze the operational requirements of Workload 1.
Workload 1 is offline batch processing, stateless, checkpoint-enabled, and fault-tolerant.
Fault-tolerant workloads that can withstand instance termination are ideal candidates for Spot VMs, which offer up to 60-91% discounts compared to standard VM pricing.
2
Analyze the operational requirements of Workload 2.
Workload 2 requires high RAM density per vCPU and continuous, uninterrupted 24/7 uptime over a predictable multi-year lifecycle.
Memory-optimized machine families (e.g., M1/M2/M3) cater to RAM-intensive database engines. Committing to a 1-year or 3-year term via Committed Use Discounts yields much higher discounts than relying solely on automatic Sustained Use Discounts.
3
Evaluate and eliminate non-viable options.
Discard strategies recommending Spot VMs for persistent databases or relying on automatic Sustained Use Discounts for long-term predictable commitments.
Spot VMs risk data disruption for non-fault-tolerant databases, and Sustained Use Discounts yield lower savings compared to contractual Committed Use Discounts for predictable multi-year workloads.

Key Concept

Selecting machine families and pricing models based on fault tolerance, memory density, and uptime predictability.
Estimated Time:2m 0s
Rate this question

Topics

Planning Compute Engine Resources and Machine Types

More questions from this topic