Examkin

Question

Difficulty: MediumCost-Optimized Compute Selection and Purchasing Strategies

A logistics and supply chain management company is migrating its operations platform to AWS. The platform consists of three main workloads:

1. A fleet tracking microservice running on Amazon ECS on AWS Fargate. This service runs 24/7 and has a highly predictable baseline CPU and memory utilization.
2. A route optimization engine running on Amazon EC2 instances that is executed once every night. The optimization engine takes approximately 3 hours to complete, is highly fault-tolerant, can be paused and restarted from checkpoints, and requires significant compute capacity.
3. An inventory database tier running on Amazon RDS for PostgreSQL that operates 24/7.

Which combination of purchasing strategies should the solutions architect recommend to achieve the most cost-effective compute architecture? (Select TWO.)

  1. Purchase a Compute Savings Plan to cover the baseline usage of the fleet tracking microservice running on AWS Fargate.Answer
  2. Use Amazon EC2 Spot Instances to run the nightly route optimization engine.Answer
  3. C
    Purchase a Compute Savings Plan to cover the continuous compute costs of the Amazon RDS for PostgreSQL database tier.
  4. D
    Migrate the fleet tracking microservice from AWS Fargate to AWS Lambda to run continuously 24/7.
  5. E
    Purchase EC2 Instance Savings Plans to cover the baseline usage of the Amazon RDS for PostgreSQL database instances.

Answer

The correct options are to purchase a Compute Savings Plan to cover the baseline usage of the fleet tracking microservice on AWS Fargate, and to use Amazon EC2 Spot Instances to run the nightly route optimization engine.
Purchasing a Compute Savings Plan is correct because it offers the flexibility to apply discounts across EC2, Lambda, and AWS Fargate, making it perfect for the 24/7 predictable fleet tracking microservice. Using Spot Instances is correct because the nightly route optimization engine is highly fault-tolerant and can restart from checkpoints, allowing it to take advantage of the deeply discounted Spot pricing without risking task failure.

Step-by-Step Solution

1
Analyze the fleet tracking microservice requirements.
It runs 24/7 on AWS Fargate with a predictable baseline.
Identify the most suitable purchasing model for predictable Fargate compute, which is a Compute Savings Plan.
2
Analyze the route optimization engine requirements.
It runs nightly for 3 hours, requires high compute capacity, and is fault-tolerant and checkpointed.
Identify the most suitable purchasing model for interruptible, non-continuous batch compute, which is EC2 Spot Instances.
3
Analyze the RDS database tier requirements.
It runs 24/7 on Amazon RDS for PostgreSQL.
Determine if Savings Plans apply to RDS. Since they do not, rule out options suggesting Compute Savings Plans or EC2 Instance Savings Plans for RDS, and recognize that RDS Reserved Instances would be needed instead.

Key Concept

Selecting cost-optimized purchasing strategies (Savings Plans, Spot Instances) for different compute hosting models (ECS Fargate, EC2) while understanding their scope limitations.
Rate this question

Topics

Cost-Optimized Compute Selection and Purchasing Strategies