A retail analytics company processes point-of-sale (POS) data uploaded by thousands of sto… — AWS Certified Solutions Architect - Professional

A retail analytics company processes point-of-sale (POS) data uploaded by thousands of stores. The processing workload runs on a fleet of Amazon EC2 instances in private subnets of a VPC in the `eu-west-1` Region. The EC2 instances retrieve approximately $250\text{ TB}$ of raw data monthly from an Amazon S3 bucket in `eu-west-1` and write $150\text{ TB}$ of processed transactional summaries to an Amazon DynamoDB table in the same Region. Currently, all outbound traffic from the private subnets is routed through NAT Gateways deployed across two Availability Zones, resulting in high data processing charges.

Additionally, the raw POS data in the S3 bucket is actively analyzed for the first $30$ days. After $30$ days, the data is rarely accessed but must be retained for $7$ years to comply with regulatory audits. The company requires audit retrievals to complete in less than $5$ hours.

Which two actions should a Solutions Architect recommend to minimize storage and data transfer costs while maintaining application availability and meeting regulatory requirements? (Select two.)

Create Gateway VPC Endpoints for Amazon S3 and Amazon DynamoDB in the VPC, and update the route tables of the private subnets to route traffic to these endpoints.Answer
Configure an Amazon S3 Lifecycle policy on the S3 bucket to transition the raw POS data to the Amazon S3 Glacier Flexible Retrieval tier $30$ days after creation.Answer
C
Route all outbound traffic from the private subnets through a single NAT Gateway in one Availability Zone to minimize the hourly NAT Gateway provisioning costs.
D
Utilize AWS Snowcone devices to physically transfer the monthly $250\text{ TB}$ of raw POS data from the stores directly to AWS to eliminate internet ingestion costs.
E
Purchase an EC2 Instance Savings Plan to offset both the EC2 instance compute costs and the data transfer processing charges incurred by the NAT Gateways.

Answer

Create Gateway VPC Endpoints for Amazon S3 and Amazon DynamoDB, and configure an Amazon S3 Lifecycle policy to transition raw POS data to the Amazon S3 Glacier Flexible Retrieval tier 30 days after creation.

Creating Gateway VPC Endpoints for Amazon S3 and Amazon DynamoDB allows the private EC2 instances to communicate directly with these services over the AWS network without traversing the NAT Gateways. Because Gateway Endpoints are free and do not incur data processing fees, this eliminates the data transfer costs associated with the

250\text{ TB}

of S3 traffic and

150\text{ TB}

of DynamoDB traffic. Transitioning raw POS data to Amazon S3 Glacier Flexible Retrieval after

30

days satisfies the regulatory retention period of

7

years at a significantly lower storage cost than S3 Standard, while meeting the retrieval time SLA of less than

5

hours (since standard retrieval for Glacier Flexible Retrieval is

3\text{ to }5\text{ hours}

Step-by-Step Solution

Analyze the current data transfer pathways and identify the source of the high costs.

Traffic to Amazon S3 (

250\text{ TB}

) and Amazon DynamoDB (

150\text{ TB}

) passes through NAT Gateways, incurring data processing charges of

\$0.045\text{ per GB}

Identifying the data volumes and destination services helps target the most cost-effective VPC endpoint type.

Select the correct endpoint solution for the target services.

Deploying Gateway VPC Endpoints for Amazon S3 and Amazon DynamoDB is free of charge and keeps traffic within the AWS network, bypassing the NAT Gateways.

Gateway VPC Endpoints eliminate NAT Gateway data processing charges for traffic destined to S3 and DynamoDB without adding hourly endpoint fees.

Evaluate S3 storage lifecycle and access requirements.

The raw POS data (

250\text{ TB}

/month) is only actively used for

30

days but must be kept for

7

years, with retrievals needed in less than

5

hours.

Transitioning data to a cheaper storage class after

30

days will reduce storage costs significantly.

Select the correct S3 storage class for the transition.

Amazon S3 Glacier Flexible Retrieval offers standard retrieval times of

3\text{ to }5\text{ hours}

and is much cheaper than S3 Standard, while S3 Glacier Deep Archive has retrieval times up to

12\text{ hours}

and cannot meet the limit.

Choosing S3 Glacier Flexible Retrieval meets the audit performance SLA of less than

5

hours while maximizing cost savings.

Key Concept

VPC Gateway Endpoints and S3 Lifecycle policies are primary tools for reducing AWS data transfer and storage costs without compromising availability or performance SLAs.

Estimated Time:3m 0s

Question

Answer

Step-by-Step Solution

Key Concept

Common Mistakes

Topics