A company is designing a cost-optimization strategy for its inventory management system on AWS. The workload consists of a containerized application layer running on Amazon ECS with AWS Fargate that requires a continuous, steady-state baseline of compute capacity, and a database tier running on Amazon RDS for PostgreSQL that also requires steady-state baseline compute and storage capacity. The company wants to maximize cost discounts for both tiers while maintaining the existing architecture and minimizing operational overhead. Which combination of purchasing strategies should a solutions architect recommend to meet these requirements most cost-effectively?
- APurchase an EC2 Instance Savings Plan to cover both the AWS Fargate baseline compute capacity and the Amazon RDS database instances.
- BMigrate the containerized application layer to AWS Lambda to run the continuous baseline tasks, and purchase Amazon RDS Reserved Instances for the database tier.
- Purchase a Compute Savings Plan to cover the AWS Fargate baseline compute capacity, and purchase Amazon RDS Reserved Instances for the database tier.Cevap
- DPurchase a Compute Savings Plan to cover both the AWS Fargate baseline compute capacity and the Amazon RDS database instances.
Cevap
Purchase a Compute Savings Plan to cover the AWS Fargate baseline compute capacity, and purchase Amazon RDS Reserved Instances for the database tier.
The correct option is to purchase a Compute Savings Plan to cover the AWS Fargate baseline compute capacity and purchase Amazon RDS Reserved Instances for the database tier. Compute Savings Plans automatically apply to Fargate compute usage, offering significant savings for steady-state workloads. Since Compute Savings Plans do not extend to database services like Amazon RDS, purchasing RDS Reserved Instances is the correct and most cost-effective method to cover the steady-state database tier.
Adım Adım Çözüm
Anahtar Kavram
Selecting and combining AWS compute purchasing models (Compute Savings Plans and RDS Reserved Instances) based on service scope and workload steadiness.