A company is designing a 3-year cost-optimization strategy for a web application deployed … — AWS Certified Solutions Architect - Associate

A company is designing a $3$ -year cost-optimization strategy for a web application deployed on AWS. The application consists of the following components:
- A containerized API hosted on Amazon ECS with AWS Fargate that requires a steady-state baseline of $4\text{ vCPUs}$ and $8\text{ GB}$ of RAM. During peak hours, the API dynamically scales up to $16\text{ vCPUs}$ and $32\text{ GB}$ of RAM.
- A batch processing job that runs once daily for $3\text{ hours}$ . The job is fault-tolerant, can be interrupted without loss of progress, and requires a minimum of $8\text{ vCPUs}$ .
- An Amazon DynamoDB table that stores user sessions. The table experiences highly unpredictable, spiky traffic patterns with long periods of inactive idle time.

Which combination of compute purchasing models and database capacity settings will provide the most cost-effective architecture?

A
Purchase a $3$ -year EC2 Instance Savings Plan to cover the baseline Fargate capacity. Use Fargate Spot for the scaling tasks. Run the daily batch processing job on Amazon EC2 On-Demand Instances, and configure the DynamoDB table to use Provisioned capacity mode with high static write capacity.
B
Purchase a $3$ -year Compute Savings Plan to cover both the baseline Fargate capacity and the DynamoDB read/write capacity. Use Fargate Spot for the scaling tasks. Run the daily batch processing job on Amazon EC2 Spot Instances, and configure the DynamoDB table to use Provisioned capacity mode with Auto Scaling.
Purchase a $3$ -year Compute Savings Plan to cover the baseline Fargate capacity. Use Fargate Spot for the scaling tasks. Run the daily batch processing job on Amazon EC2 Spot Instances, and configure the DynamoDB table to use On-Demand capacity mode.Cevap
D
Purchase a $3$ -year Compute Savings Plan to cover the baseline Fargate capacity. Use On-Demand Fargate tasks for the scaling tasks. Re-architect the daily batch processing job to run on AWS Lambda, and configure the DynamoDB table to use On-Demand capacity mode.

Cevap

The correct answer is the option that purchases a

3

-year Compute Savings Plan for the baseline Fargate capacity, uses Fargate Spot for scaling Fargate tasks, runs the daily batch processing job on Amazon EC2 Spot Instances, and configures the DynamoDB table to use On-Demand capacity mode.

The correct answer is correct because a

3

-year Compute Savings Plan provides the highest discount for the steady-state baseline Fargate tasks, which run continuously. Using Fargate Spot for the scaling API capacity allows the application to scale cost-effectively. Running the

3

-hour fault-tolerant batch job on Amazon EC2 Spot Instances leverages the lowest-cost compute option for interruptible workloads, while avoiding the

15\text{-minute}

execution limit of AWS Lambda. Finally, configuring Amazon DynamoDB in On-Demand capacity mode is optimal for spiky, unpredictable traffic with long idle periods, as it avoids paying for unused provisioned capacity.

Adım Adım Çözüm

Analyze the baseline compute requirement for the user-facing API.

The baseline ECS Fargate capacity runs continuously (

24/7

) and is best covered by a

3

-year Compute Savings Plan, which offers significant discounts for Fargate workloads.

Steady-state workloads running continuously benefit most from long-term commitments like Savings Plans.

Analyze the scaling compute requirement for the user-facing API.

The scaling Fargate capacity is transient and dynamic, making Fargate Spot the most cost-effective option to handle scale-out events.

Using Spot capacity for non-critical, scalable compute tiers minimizes On-Demand costs.

Analyze the batch processing job requirement.

Since the job is fault-tolerant, runs for

3\text{ hours}

, and can tolerate interruptions, Amazon EC2 Spot Instances should be selected.

AWS Lambda cannot be used because its maximum execution timeout is

15\text{ minutes}

, and Spot Instances provide up to a

90\%

discount compared to On-Demand compute.

Analyze the database capacity requirement.

The DynamoDB table experiences highly unpredictable, spiky traffic with long idle periods, which aligns with On-Demand capacity mode.

On-Demand capacity mode charges per request and scales to zero during idle periods, avoiding the cost of over-provisioned capacity.

Anahtar Kavram

Cost-optimizing AWS compute requires matching workloads to appropriate purchasing models: Compute Savings Plans for steady-state Fargate/EC2 baseline compute, Spot capacity (Fargate Spot or EC2 Spot) for interruptible and scaling components, and On-Demand database capacity modes for spiky, unpredictable access patterns.

Tahmini Süre:2m 30s

Soru

Cevap

Adım Adım Çözüm

Anahtar Kavram

Sık Yapılan Hatalar

Konular