Examkin

Question

Difficulty: MediumCost-Optimized Compute Selection and Purchasing Strategies

A company runs a telemetry ingestion and processing platform on AWS. The workload has the following components:
- A baseline of four Amazon EC2 instances that run continuously 24/7 to receive telemetry data.
- An Auto Scaling group of EC2 instances that adds between two and eight additional instances during peak daytime hours.
- A daily batch analytics job running on Amazon EC2 for two hours that is fault-tolerant and can be interrupted or resumed.
- An Amazon RDS PostgreSQL database instance that runs continuously 24/7 to store the processed telemetry.

Which combination of purchasing options will provide the most cost-effective solution for this workload?

  1. Purchase a Compute Savings Plan to commit to the baseline EC2 usage, use On-Demand Instances for the peak scaling instances, use Spot Instances for the batch analytics job, and purchase an RDS Reserved Instance for the database.Answer
  2. B
    Purchase a Compute Savings Plan to commit to both the baseline EC2 usage and the Amazon RDS database usage, use On-Demand Instances for the peak scaling instances, and use Spot Instances for the batch analytics job.
  3. C
    Purchase Instance Savings Plans to cover both the baseline and peak scaling EC2 instances, use On-Demand Instances for the batch analytics job, and purchase an RDS Reserved Instance for the database.
  4. D
    Use Spot Instances for the baseline EC2 instances and the batch analytics job, use On-Demand Instances for the peak scaling instances, and purchase a Compute Savings Plan to cover the Amazon RDS database.

Answer

Purchase a Compute Savings Plan to commit to the baseline EC2 usage, use On-Demand Instances for the peak scaling instances, use Spot Instances for the batch analytics job, and purchase an RDS Reserved Instance for the database.
The correct strategy combines a Compute Savings Plan for the baseline EC2 workload, On-Demand Instances for the variable peak scaling compute, Spot Instances for the short-term and fault-tolerant batch processing, and an RDS Reserved Instance to discount the database layer. This maximizes discounts while avoiding over-commitment and maintaining application availability.

Step-by-Step Solution

1
Analyze the 24/7 baseline compute requirements.
The baseline of four EC2 instances running 24/7 represents a predictable, continuous compute workload suitable for a commitment-based discount model like a Compute Savings Plan.
Savings Plans offer significant discounts over On-Demand rates in exchange for a commitment to a consistent amount of usage.
2
Evaluate the peak scaling and batch compute requirements.
The peak scaling instances (2 to 8 instances) are unpredictable and short-lived, making them best suited for On-Demand Instances to avoid paying for idle committed capacity. The daily batch analytics job is fault-tolerant and runs for only 2 hours, making it ideal for Spot Instances.
Spot Instances offer up to a 90% discount compared to On-Demand instances, with the trade-off of potential interruption, which is acceptable for fault-tolerant batch jobs.
3
Determine the purchasing option for the database tier.
The Amazon RDS PostgreSQL instance runs continuously 24/7. Since Compute Savings Plans do not cover RDS, an RDS Reserved Instance must be purchased to optimize costs.
RDS Reserved Instances provide a discount for database instances with a one- or three-year commitment.

Key Concept

Selecting cost-optimized compute purchasing models based on workload predictability, fault tolerance, and service support boundaries.
Rate this question

Topics

Cost-Optimized Compute Selection and Purchasing Strategies