Question

Difficulty: MediumOptimizing Resource Costs and Sizing

A financial services company runs a multi-tier application across a multi-account structure in AWS Organizations. The application consists of a web layer running on AWS Fargate in private subnets across two Availability Zones (AZs), a batch processing layer on Amazon EC2 m6i.large instances, and a reporting API built with AWS Lambda. To minimize costs, the operations team proposes deploying a single NAT Gateway in one AZ to route outbound internet traffic from the private subnets. Additionally, the company plans to share specialized network resources via AWS Resource Access Manager (RAM) to a partner's AWS account that is outside their AWS Organization. The company wants to purchase Savings Plans to reduce compute costs for EC2, Fargate, and Lambda, while keeping maximum flexibility for future microservice migrations. Which combination of actions should a Solutions Architect recommend to optimize costs, maintain high availability, and ensure successful resource sharing?

  1. A
    Purchase EC2 Instance Savings Plans for the m6i instance family at the AWS Organizations management account level. Deploy a NAT Gateway in each Availability Zone. Enable sharing with external accounts in the AWS Resource Access Manager settings to share resources with the partner account.
  2. Purchase Compute Savings Plans at the AWS Organizations management account level. Deploy a NAT Gateway in each Availability Zone. Enable sharing with external accounts in the AWS Resource Access Manager settings to share resources with the partner account.Answer
  3. C
    Purchase Compute Savings Plans at the AWS Organizations management account level. Keep the single NAT Gateway configuration to minimize NAT Gateway hourly fees and data processing costs. Enable sharing with external accounts in the AWS Resource Access Manager settings to share resources with the partner account.
  4. D
    Purchase Compute Savings Plans at the AWS Organizations management account level. Deploy a NAT Gateway in each Availability Zone. Share the resources using AWS Resource Access Manager without enabling external account sharing, relying on cross-account IAM trust relationships.

Answer

Purchase Compute Savings Plans at the AWS Organizations management account level, deploy a NAT Gateway in each Availability Zone, and enable sharing with external accounts in the AWS Resource Access Manager settings.
The correct option optimizes compute costs by purchasing Compute Savings Plans, which provide up to 66% savings and apply automatically to EC2, Fargate, and Lambda across all accounts in the AWS Organization. It also ensures high availability by deploying a NAT Gateway in each Availability Zone, preventing a single point of failure for outbound traffic. Finally, it enables resource sharing with a partner account outside the AWS Organization by turning on the external sharing setting in AWS Resource Access Manager.

Step-by-Step Solution

1
Determine the appropriate Savings Plan type to optimize the compute resources.
Compute Savings Plans are selected.
Compute Savings Plans apply to EC2, Fargate, and Lambda, providing the required flexibility and cost coverage for all three compute types, whereas EC2 Instance Savings Plans do not cover Fargate and Lambda.
2
Evaluate the network architecture for cost and high availability.
Deploy a NAT Gateway in each Availability Zone instead of a single NAT Gateway.
A single NAT Gateway creates a single point of failure for resources in other Availability Zones, violating the high availability constraint.
3
Determine the configuration needed to share resources via AWS RAM to an external account.
Enable sharing with external accounts in the AWS Resource Access Manager settings.
AWS RAM restricts sharing outside of the AWS Organization by default; explicit permission to share with external entities is required.

Key Concept

Balancing compute cost optimization using flexible Savings Plans with high availability networking constraints and multi-account resource sharing rules.
Rate this question