How is content referenced in Swarm?

When content is written to Swarm, the client receives a 32-digit hexadecimal number called a UUID. The UUID is a 128-bit, globally unique identifier for the content. In order to read the content from the cluster, a client contacts any node in the cluster (the primary access node or PAN) with the UUID. The PAN will direct the client to the best node (SAN- secondary access node) from which to read the content.

If the author of the content writes named streams, instead of receiving a random UUID associated with the content, the content is referenced by the name of the object (including the domain, bucket, and stream name).