A logistics company runs a distributed fleet management system across an AWS Organization … — AWS Certified Solutions Architect - Professional

A logistics company runs a distributed fleet management system across an AWS Organization with consolidated billing. The system is split into two AWS member accounts: Account A (Ingestion) runs a dynamic processing workload on Amazon EC2 m6i.xlarge instances (average CPU utilization $25\%$ , peak $40\%$ , memory usage consistently under $30\%$ ) and AWS Lambda functions for real-time telemetry processing. Account B (Analytics) hosts a cluster of Amazon ECS tasks running on AWS Fargate that processes reports. Outbound traffic to the internet and AWS services from Account A is routed through redundant NAT Gateways deployed across three Availability Zones. The company plans to optimize compute costs over a $3\text{-year}$ term. Performance must not be compromised, and no application downtime is permitted. Which optimization strategy will achieve the highest cost reduction while satisfying these requirements?

A
Consolidate the redundant NAT Gateways in Account A into a single NAT Gateway in one Availability Zone to eliminate hourly NAT Gateway charges. Right-size the EC2 instances in Account A to m6i.large, and purchase a $3\text{-year}$ Compute Savings Plan in the organization's management account.
Right-size the EC2 instances in Account A to m6i.large. Purchase a $3\text{-year}$ Compute Savings Plan in the organization's management account to cover the right-sized EC2 instances, ECS Fargate tasks, and Lambda functions across both accounts. Maintain the redundant NAT Gateways.Answer
C
Right-size the EC2 instances in Account A to m6i.large. Purchase a $3\text{-year}$ EC2 Instance Savings Plan in the organization's management account to cover the EC2 instances, Fargate tasks, and Lambda functions across both accounts. Maintain the redundant NAT Gateways.
D
Right-size the EC2 instances in Account A to m6i.large. Purchase a $3\text{-year}$ EC2 Instance Savings Plan in Account A, and use AWS Resource Access Manager (RAM) to share the savings benefits with Account B to cover the Fargate and Lambda workloads. Maintain the redundant NAT Gateways.

Answer

Right-size the EC2 instances in Account A to m6i.large, purchase a 3-year Compute Savings Plan in the organization's management account to cover the EC2 instances, ECS Fargate tasks, and Lambda functions across both accounts, and maintain the redundant NAT Gateways.

The correct strategy combines right-sizing the m6i.xlarge instances to m6i.large (since the peak workload of

1.6\text{ vCPUs}

and

4.8\text{ GiB}

of RAM easily fits within a large instance's

2\text{ vCPUs}

and

8\text{ GiB}

capacity) with purchasing a Compute Savings Plan in the management account. The Compute Savings Plan is the only plan type that applies to EC2, Fargate, and Lambda workloads across different member accounts. Retaining the redundant NAT Gateways is necessary to avoid introducing a single point of failure for outbound traffic.

Step-by-Step Solution

Analyze the EC2 CPU and memory usage to right-size the instances.

The current m6i.xlarge instances have

4\text{ vCPUs}

and

16\text{ GiB}

of RAM. Peak CPU is

40\%

(equivalent to

1.6\text{ vCPUs}

) and memory is consistently under

30\%

(under

4.8\text{ GiB}

). Downsizing to m6i.large (

2\text{ vCPUs}

and

8\text{ GiB}

of RAM) safely accommodates the peak CPU and memory demands.

Ensures that the instances are not over-provisioned while keeping performance within safe boundaries.

Select the correct Savings Plan type for the mixed compute environment.

Compute Savings Plans cover EC2, AWS Fargate, and AWS Lambda workloads across all accounts under consolidated billing. EC2 Instance Savings Plans do not cover Fargate or Lambda.

Allows the company to apply discounts to all three compute types (EC2, Fargate, and Lambda) across both member accounts.

Verify high-availability constraints for outbound network components.

Consolidating the three NAT Gateways into a single NAT Gateway in one AZ would introduce a single point of failure (SPOF) for outbound connectivity, which violates the requirement for zero application downtime. The redundant NAT Gateways must remain.

Ensures the architecture meets the high availability and zero-downtime SLA constraints.

Key Concept

Compute sizing optimization, Compute Savings Plans coverage, and high-availability architecture constraints.

Question

Answer

Step-by-Step Solution

Key Concept

Common Mistakes

Topics

More questions from this topic