Examkin

Question

Difficulty: HardPlanning Compute Engine Resources and Machine Types

An engineering team is planning Google Cloud infrastructure for two new workloads with distinct operational characteristics:

1. Workload 1: A stateful, single-node transactional database service that runs continuously 24/724/7 and requires high single-core performance with guaranteed compute uptime.
2. Workload 2: A stateless, fault-tolerant batch analysis service that processes asynchronous messages from Cloud Pub/Sub and can resume interrupted tasks automatically.

Which TWO provisioning and machine type strategies align with Google Cloud recommended practices for these workloads? (Select TWO answers.)

  1. Provision Workload 1 using standard virtual machines with Compute-Optimized (C2) or General-Purpose (N2) machine types and leverage Committed Use Discounts (CUDs).Answer
  2. B
    Provision Workload 1 on Spot VMs with automatic restart enabled to maximize cost savings for the continuous stateful service.
  3. Provision Workload 2 on Spot VMs within a Managed Instance Group (MIG) to significantly reduce batch compute costs.Answer
  4. D
    Migrate Workload 2 to Cloud Functions while hardcoding non-standard custom container port bindings to bypass VM machine type selection completely.

Answer

The team should provision Workload 1 using standard virtual machines with C2 or N2 machine types backed by Committed Use Discounts, and provision Workload 2 using Spot VMs within a Managed Instance Group.
Stateful 24/7 workloads require standard VM instances to guarantee availability, live migration support, and persistent attached storage durability; purchasing Committed Use Discounts optimizes long-term baseline cost. Conversely, stateless batch processing services that pull from decoupled queues are resilient to sudden interruptions and are ideal for Spot VMs, which provide deep discounts without risking data loss.

Step-by-Step Solution

1
Analyze Workload 1 characteristics
Identified as a stateful, continuous 24/7 database requiring guaranteed single-core performance.
Stateful databases require high availability, live migration capabilities, and predictable performance, making standard VMs with Committed Use Discounts optimal.
2
Analyze Workload 2 characteristics
Identified as a stateless, fault-tolerant batch processing service.
Fault-tolerant batch processing can handle instance preemptions gracefully, allowing the use of heavily discounted Spot VMs in a Managed Instance Group.
3
Evaluate candidate strategies against Google Cloud best practices
Selected standard VMs with CUDs for Workload 1 and Spot VMs in a MIG for Workload 2.
Matches workload requirements precisely with Google Cloud cost optimization and reliability patterns.

Key Concept

Compute Engine Machine Type Selection and Spot VM vs. Committed Use Discount Suitability
Rate this question

Topics

Planning Compute Engine Resources and Machine Types

More questions from this topic