A method and system of optimizing content delivery at a processing element in a dynamic content delivery architecture, the method having the steps of: receiving a content and metadata envelope at the processing element; checking the content and metadata envelope to determine whether the content and metadata envelope includes metadata for the processing element; if the content envelope contains metadata for the processing element, extracting and caching the metadata; if the content envelope does not contain metadata for the processing element, retrieving metadata for a content provider associated with the content from a cache on the processing element; and applying the extracted or retrieved metadata to the content and metadata envelope.