Programming Amazon Web Services

Alternative Hostnames

S3 offers two mechanisms that allow you to use alternative hostnames to access the contents of your buckets. The first mechanism, called Virtual Hosting, allows you to use your own domain name as an alias for an S3 bucket. This feature makes the content distribution capabilities of S3 much more attractive, because you can provide your content through a domain name of your choice, while still serving data directly from S3. The second mechanism allows you to access your buckets via a subdomain of the S3 service domain.

To support alternative hostnames, S3 interprets the Host header in HTTP request messages to deduce the name of the bucket the request is referring to. Conveniently, the vast majority of HTTP client applications, including all web browsers, supply the Host header automatically when they make a request. Because the bucket name can be determined from the alternative hostnames, there is no need to include the bucket’s name in the URI path for requests that use alternative hosts.

The following table demonstrates the different URIs that can be used to refer to the same object in S3 using different hostnames. The first request uses a standard S3 URI to refer to an object, and it includes the bucket name in the URI’s path. The second request uses an S3 subdomain, in which case the hostname starts with the bucket’s name. The third uses a Virtual Host domain name, and the bucket name is represented by the entire hostname (this domain does not actually exist). ...