Examkin

Question

Difficulty: HardPlanning Compute Engine Resources and Machine Types

A renewable energy enterprise is planning the Google Cloud Compute Engine architecture for two distinct operational workloads:
1. A primary grid telemetry routing engine that runs continuously 24/7, handles live stateful transactions, and must maintain a strict uptime SLA without unexpected termination.
2. An asynchronous nightly batch data processing pipeline that converts raw sensor logs into aggregated daily analytics; the pipeline supports native application checkpointing and can resume execution seamlessly if interrupted.

You need to design a Compute Engine deployment strategy that minimizes overall operational cost while adhering to the SLA and fault-tolerance requirements of each workload. Which TWO recommendations should you implement? (Select TWO.)

  1. Deploy the 24/7 grid telemetry routing engine on standard Compute Engine virtual machines and purchase Committed Use Discounts (CUDs) for the baseline capacity.Answer
  2. Provision the nightly batch data processing pipeline using Spot Virtual Machines (VMs) to take advantage of deeply discounted compute pricing for fault-tolerant tasks.Answer
  3. C
    Deploy the continuous 24/7 grid telemetry routing engine on Spot Virtual Machines to achieve maximum hourly cost savings for stateful continuous operations.
  4. D
    Configure the nightly batch analytics process on dedicated sole-tenant nodes with 3-year Committed Use Discounts to prevent instance preemption during batch execution.

Answer

The correct recommendations are to deploy the 24/7 grid telemetry engine on standard virtual machines using Committed Use Discounts for predictable baseline capacity, and to run the nightly checkpoint-supported batch process on Spot Virtual Machines to maximize cost efficiency.
For continuous 24/7 stateful workloads with strict SLAs, standard VM instances backed by Committed Use Discounts provide guaranteed capacity and significant cost reduction. For fault-tolerant batch workloads that support checkpointing, Spot VMs provide deep cost discounts while allowing the system to handle any potential instance preemptions gracefully.

Step-by-Step Solution

1
Analyze the availability requirement of the 24/7 stateful telemetry routing engine.
The continuous SLA and stateful nature require non-interruptible standard instances.
Interruptible instances (Spot VMs) risk breaking uptime SLAs due to sudden preemption.
2
Select the cost-optimization mechanism for the 24/7 baseline continuous workload.
Apply Committed Use Discounts (CUDs) for 1 or 3 years.
CUDs significantly reduce cost for steady-state compute workloads without affecting VM availability.
3
Evaluate the fault-tolerance characteristics of the nightly batch analytics process.
The workload supports native checkpointing and asynchronous retry logic.
Since state can be recovered after interruption, Spot VMs are ideal for minimizing compute expenditure.

Key Concept

Compute Engine Workload Planning: Spot VMs for Fault-Tolerant Batch Workloads vs Committed Use Discounts for Steady-State 24/7 Workloads
Rate this question

Topics

Planning Compute Engine Resources and Machine Types

More questions from this topic