Interesting concept in AWS. You can share your data in AWS S3 but not pay for storing this data. In AWS it’s called “Requester Pays Buckets”.

https://docs.aws.amazon.com/AmazonS3/latest/userguide/RequesterPaysBuckets.html

https://docs.aws.amazon.com/AmazonS3/latest/userguide/ObjectsinRequesterPaysBuckets.html

AS result for data transfer and storing will pay requester who will download these data.

It is very useful approach to store datasets in AWS S3.

How to download data with S3 CLI

For example for arXiv.org shared data it will be:

aws s3 cp s3://arxiv/pdf/arXiv_pdf_manifest.json . --request-payer requester --profile personal

Access through http

• For GET, HEAD, and POST requests, include x-amz-request-payer : requester in the header
• For signed URLs, include x-amz-request-payer=requester in the request