SAA-C03 Practice Question 69: S3 Requester Pays Dataset
AWS Certified Solutions Architect – Associate (SAA-C03) · Test 7 · question 9 of 10
A research institute publishes a 500 TB climate dataset in S3 and lets any AWS customer download it. The institute is happy to keep paying to store the data, but the request and data transfer charges generated by downloaders have grown into the largest line on its bill. It wants the people pulling the data to carry that cost, while the dataset stays openly available.
What should the institute do?
Show answer and explanation
Answer:
Requester Pays moves exactly the two charges in question. With it enabled, the requester rather than the bucket owner pays for the request and for the data download, while the bucket owner continues to pay for storage, which matches what the institute is willing to fund. AWS names publishing large datasets as the typical reason to use it. The one behavioural change is that anonymous access is no longer permitted: every request must be authenticated so that S3 knows whom to bill, and requesters confirm they accept the charge with the x-amz-request-payer header.
Option A changes the storage class, not who pays. Retrieval from an archive tier is billed to the bucket owner, so the institute would be adding retrieval fees and hours of restore latency on top of a transfer bill it still pays.
Option B relocates the cost rather than shifting it. Caching cuts origin transfer, but CloudFront data transfer out to the internet is billed to the distribution's owner, still the institute, and a 500 TB dataset downloaded in bulk caches poorly.
Option C is an access control mechanism. Presigned URLs govern who may download and for how long; every byte served through one is still charged to the bucket owner.
Reference: Using Requester Pays general purpose buckets for storage transfers and usage