Examkin

Question

Difficulty: MediumPlanning Compute Engine Resources and Machine Types

An enterprise software company is planning the Google Compute Engine infrastructure for a mission-critical, self-hosted in-memory cache running continuously 24/7 with predictable, high-memory resource requirements guaranteed for the next three years. The workload requires high availability and cannot tolerate unexpected VM preemptions or terminations. Which Compute Engine machine configuration and purchasing strategy should the cloud engineer select?

  1. Provision memory-optimized or high-memory standard VM instances and purchase a 3-year Committed Use Discount (CUD) for the baseline compute capacity.Answer
  2. B
    Provision Spot VM instances using memory-optimized machine types to minimize compute costs for the caching layer.
  3. C
    Provision general-purpose VM instances relying solely on automatic Sustained Use Discounts (SUDs) to achieve maximum long-term savings over the 3-year period.
  4. D
    Deploy the in-memory cache workload as a Google Cloud Functions service to eliminate VM server management overhead.

Answer

Provision memory-optimized or high-memory standard Compute Engine instances paired with a 3-year Committed Use Discount (CUD) commitment.
Memory-optimized or high-memory Compute Engine instance families provide tailored vCPU-to-memory ratios for RAM-heavy workloads such as in-memory caches. Securing capacity with 3-year Committed Use Discounts yields maximum cost optimization for steady, uninterrupted 24/7 baseline workloads without risking VM preemption.

Step-by-Step Solution

1
Analyze workload resource profile and availability requirements.
Identified a memory-bound, stateful caching workload running 24/7 that cannot tolerate interruptions.
Choosing the correct machine family requires matching high RAM demands while ruling out preemptible/Spot VMs due to zero tolerance for unexpected shutdowns.
2
Evaluate machine family selection.
Selected memory-optimized or high-memory machine configurations (such as M2/M3 or N2/E2 high-mem).
These machine families deliver optimal cost-per-GB of RAM for memory-intensive caching systems.
3
Evaluate pricing and commitment options for predictable continuous usage.
Selected 3-year Committed Use Discounts (CUDs).
For continuous 24/7 baseline capacity over 3 years, CUDs provide deep discounts (up to 57%-70%) compared to standard on-demand pricing or automatic Sustained Use Discounts.

Key Concept

Compute Engine Machine Family Selection and Committed Use Discounts (CUDs)
Estimated Time:1m 15s
Rate this question

Topics

Planning Compute Engine Resources and Machine Types

More questions from this topic