I created a new (not even a week old) public S3-bucket to access some files remotely (it has to be public - that is the crux). Things like logging and versioning are deactivated, and pretty much all the standard options were kept.
However, I have a huge amount of requests per day (over 3000requests per day and maybe even increasing) that is not connected at all to my assessing of the files. Where does such a traffic come from? Does amazon access the files itself by default? Something like a tracking of the files? Can this be deactivated?
You can activate Amazon S3 server access logging to obtain this information. The logs will show the requests and their origin.
FYI, requests are charged at $0.0004 per 1000, therefore your 3000 requests per day are costing 1.2c per day.