Examkin

Question

Difficulty: MediumCost-Optimized Compute Selection and Purchasing Strategies

A company is designing a cost-effective infrastructure for two new workloads and an existing database on AWS:

1. A backend application running on Amazon EC2 instances that runs continuously 24/7 with a baseline of 10 instances. Every Friday, a batch reporting job runs for 8 hours, requiring an additional 20 instances. These batch jobs are stateless, fault-tolerant, and can be resumed if interrupted.
2. A containerized microservices application running on Amazon ECS. The workload has highly unpredictable, dynamic traffic, and the company wants to minimize server provisioning and operational overhead.
3. An Amazon RDS for PostgreSQL database instance that runs continuously 24/7.

Which of the following strategies should a solutions architect select to meet these requirements at the lowest cost? (Select TWO.)

  1. Purchase a Compute Savings Plan to cover the baseline 10 EC2 instances and the ECS Fargate tasks.Answer
  2. Launch the weekly Friday batch reporting jobs on Amazon EC2 Spot Instances.Answer
  3. C
    Purchase a Compute Savings Plan to cover the baseline 10 EC2 instances, the ECS Fargate tasks, and the Amazon RDS for PostgreSQL database.
  4. D
    Host the continuous 24/7 backend application on AWS Lambda to eliminate server provisioning and minimize operational overhead.
  5. E
    Use AWS Cost Explorer to automatically scale down the EC2 instance fleet once the Friday batch reporting jobs complete.

Answer

Purchase a Compute Savings Plan to cover the baseline 10 EC2 instances and the ECS Fargate tasks, and launch the weekly Friday batch reporting jobs on Amazon EC2 Spot Instances.
The correct strategy combines a Compute Savings Plan to cover both the baseline 24/7 EC2 instances and the serverless ECS Fargate tasks (since Compute Savings Plans flexibly apply to both compute types), along with using Spot Instances for the short-lived, stateless, and interruptible Friday reporting jobs to maximize cost savings without commitment waste.

Step-by-Step Solution

1
Analyze the baseline workload and container hosting requirements to minimize management overhead.
The baseline of 10 EC2 instances runs 24/7, and the ECS microservices need serverless scaling (AWS Fargate) to minimize operational overhead.
AWS Fargate is the ideal launch type for Amazon ECS to eliminate server management. A Compute Savings Plan covers both EC2 and Fargate usage, offering significant discounts with high flexibility.
2
Analyze the periodic weekly Friday batch reporting job requirements.
The job runs for only 8 hours a week, and is stateless, fault-tolerant, and interruptible.
Since the workload is short-lived, stateless, and tolerant to interruptions, EC2 Spot Instances are the most cost-effective selection, avoiding commitment costs of Savings Plans or the higher rates of On-Demand.
3
Evaluate the database optimization and discount coverage scope.
The Amazon RDS instance runs 24/7, but cannot be covered by Compute Savings Plans.
Compute Savings Plans only cover EC2, Fargate, and Lambda. To optimize RDS costs, RDS Reserved Instances must be purchased separately.

Key Concept

Selecting appropriate compute hosting models and matching them with optimized purchasing options (Compute Savings Plans for baseline EC2/Fargate, Spot Instances for interruptible batch workloads) while understanding coverage scopes.
Rate this question

Topics

Cost-Optimized Compute Selection and Purchasing Strategies