Examkin

Question

Difficulty: MediumPlanning Compute Engine Resources and Machine Types

A financial analytics company is designing the infrastructure on Google Cloud to support two distinct internal services. Service 1 is a non-standard stateless data transformation microservice requiring 3 vCPUs and 20 GB of memory. Service 2 is a large-scale batch simulation job that is fully fault-tolerant, stateless, and capable of resuming interrupted tasks from intermediate checkpoints. Which TWO compute resource planning strategies should you recommend to minimize compute costs while meeting operational requirements? (Select TWO.)

  1. Provision Custom Machine Types for Service 1 to match the precise vCPU and memory ratio without paying for unused predefined resources.Answer
  2. Provision Spot Virtual Machines (VMs) for Service 2 to run the fault-tolerant batch simulation workload at significant cost savings.Answer
  3. C
    Deploy the primary stateful transactional relational database supporting the simulation engine on Spot VMs to reduce baseline infrastructure costs.
  4. D
    Combine Spot VM pricing for Service 2 with automatic Sustained Use Discounts (SUDs) to compound monthly savings.

Answer

The correct recommendations are to provision Custom Machine Types for the non-standard microservice and to use Spot Virtual Machines for the fault-tolerant batch simulation job.
Custom Machine Types allow precise vCPU and memory sizing for non-standard workload ratios to avoid over-provisioning unused resources. Spot Virtual Machines deliver maximum cost savings for stateless, fault-tolerant batch workloads that can tolerate unexpected interruptions.

Step-by-Step Solution

1
Analyze the requirements for Service 1 (3 vCPUs and 20 GB RAM).
Standard machine types follow fixed vCPU-to-RAM ratios (e.g., n2-standard-4 has 4 vCPUs and 16 GB RAM). To get 20 GB RAM using predefined types, one would have to over-provision vCPUs.
Custom Machine Types permit granular allocation of vCPUs and memory tailored to specific application requirements, optimizing costs.
2
Analyze the requirements for Service 2 (fault-tolerant batch simulation job).
Since the batch job can handle interruptions and resume from checkpoints, high availability is not required.
Spot VMs offer steep discounts for compute instances that can be reclaimed by Compute Engine at any time, making them ideal for batch processing.
3
Evaluate and eliminate incorrect distractor strategies.
Stateful databases cannot risk sudden Spot terminations, and Spot VMs are not eligible for Sustained Use Discounts.
Prevents anti-patterns in Compute Engine workload planning.

Key Concept

Compute Engine Machine Types & Provisioning Models
Rate this question

Topics

Planning Compute Engine Resources and Machine Types

More questions from this topic