Cost capping Amazon S3 (Simple Storage Service)

Question

I find the fact that the service does not offer cost capping a bit scary. Are there alternatives (to Amzn s3 that offers capping), or is there a way to for example poll the current cost and disable the service if data or transfer volumes suddenly grow like crazy?

Or some other clever way of capping the bills.

Answer 1

First of all, this question is obviously relevant.

If I understand you correctly, you would like to have an estimate of your bandwith costs + some way of stop people downloading your objects.

For estimating your bandwidth costs, you have two options:

1. Download an usage report from your AWS account; if you choose Data-Transfer-Out and All-Operations there, you can easily get from that a fairly accurate estimate of your bandwidth costs. You have a considerable amount of detail available, but this is a manual process (in spite of lots of requests for an API for this at the AWS forums). The closest thing to programmatic access that I've came across is this script
2. enable bucket logging, and analyze the logs yourself, or through some external tool/service such as S3STAT.

and concerning blocking access etc,

1. change the object/s and/or bucket/s ACLs from which you're getting most of the downloads, making them private.
2. make your objects available through your own domain, and remap the relevant DNS records for those objects/buckets; you could use AWS Route 53 for this, for example.