Question

A media company runs a large data ingestion platform that uses S3 multipart uploads for large files uploaded from thousands of edge devices. Devices frequently lose connectivity mid-upload, leaving many uploads never completed. The FinOps team notices the S3 bucket's storage cost is significantly higher than the sum of the visible objects reported in the console and via ListObjects. The team wants to reduce this hidden cost with minimal operational overhead and without changing device upload logic. What should the solutions architect recommend?

Accepted Answer

Configure an S3 Lifecycle rule with an AbortIncompleteMultipartUpload action to automatically delete incomplete multipart upload parts after a set number of days.

Answer

Enable S3 Intelligent-Tiering on the bucket so that incomplete multipart upload parts transition to a lower-cost storage tier automatically.

Answer

Enable S3 Versioning and add a lifecycle rule to expire noncurrent versions after 7 days to clean up orphaned upload data.

Answer

Create a scheduled Lambda function that calls ListObjectsV2 daily and deletes any objects that have not been fully written.