AWS Certified Solutions Architect – Professional — Question 241

A company has an application that runs on a fleet of Amazon EC2 instances and stores 70 GB of device data for each instance in Amazon S3. Recently, some of the S3 uploads have been failing. At the same time, the company is seeing an unexpected increase in storage data costs. The application code cannot be modified.
What is the MOST efficient way to upload the device data to Amazon S3 while managing storage costs?

Answer options

• A. Upload device data using a multipart upload. Use the AWS CLI to list incomplete parts to address the failed S3 uploads. Enable the lifecycle policy for the incomplete multipart uploads on the S3 bucket to delete the old uploads and prevent new failed uploads from accumulating.
• B. Upload device data using S3 Transfer Acceleration. Use the AWS Management Console to address the failed S3 uploads. Use the Multi-Object Delete operation nightly to delete the old uploads.
• C. Upload device data using a multipart upload. Use the AWS Management Console to list incomplete parts to address the failed S3 uploads. Configure a lifecycle policy to archive continuously to Amazon S3 Glacier.
• D. Upload device data using S3 Transfer Acceleration. Use the AWS Management Console to list incomplete parts to address the failed S3 uploads. Enable the lifecycle policy for the incomplete multipart uploads on the S3 bucket to delete the old uploads and prevent new failed uploads from accumulating.

Correct answer: A

Explanation

Option A is correct because it efficiently manages failed uploads by using multipart uploads and allows for the automatic deletion of incomplete parts, which helps control storage costs. The other options either do not address the need for multipart uploads to mitigate failed uploads (B and D) or suggest archiving to S3 Glacier, which doesn't solve the immediate issue of failed uploads (C).