Examkin

Question

Difficulty: MediumAWS Pricing Models

A smart-grid utility company is deploying a fleet of IoT sensors that report electricity consumption data. The core ingestion server runs continuously 24/724/7 to process this real-time stream. However, once a month, the company runs a complex data aggregation and simulation job that takes exactly 22 hours to complete. This monthly simulation is critical, cannot be interrupted, and must run on a specific day. Which combination of Amazon EC2 pricing models represents the most cost-effective choice for these workloads?

  1. A
    Spot Instances for the core ingestion server, and On-Demand Instances for the monthly simulation job.
  2. Reserved Instances for the core ingestion server, and On-Demand Instances for the monthly simulation job.Answer
  3. C
    Reserved Instances for the core ingestion server, and Spot Instances for the monthly simulation job.
  4. D
    On-Demand Instances for the core ingestion server, and Dedicated Hosts for the monthly simulation job.

Answer

Reserved Instances for the core ingestion server, and On-Demand Instances for the monthly simulation job.
The correct option combines Reserved Instances for the steady-state ingestion server and On-Demand Instances for the monthly simulation job. The core ingestion server runs continuously (24/724/7), representing a predictable workload where Reserved Instances offer a significant discount. The monthly simulation is a short, critical, and non-interruptible task. On-Demand Instances provide the required reliability and capacity guarantee for this brief duration without requiring any upfront or long-term commitment.

Step-by-Step Solution

1
Analyze the core ingestion server workload requirements.
The server runs continuously (24/724/7), representing a predictable, steady-state workload.
Steady-state workloads are best suited for Reserved Instances or Savings Plans to minimize hourly costs.
2
Analyze the monthly simulation job workload requirements.
The job runs for only 22 hours once a month, is critical, and cannot tolerate interruptions.
Because it cannot tolerate interruptions, Spot Instances cannot be used. Because it runs only 22 hours a month, Reserved Instances would be highly wasteful. Therefore, On-Demand Instances are the most cost-effective option for this short-duration, critical task.
3
Combine the optimal pricing models for both workloads.
Reserved Instances for the core ingestion server, and On-Demand Instances for the monthly simulation job.
This combination ensures maximum cost efficiency while meeting the availability and reliability requirements of both workloads.

Key Concept

Selecting optimal Amazon EC2 pricing models based on workload predictability, duration, and fault tolerance.
Estimated Time:1m 30s
Rate this question

Topics

AWS Pricing Models

More questions from this topic