US Patent 11194719 Cache optimization

A system and method for management and processing of resource requests at cache server computing devices is provided. Cache server computing devices segment content into an initialization fragment for storage in memory and one or more remaining fragments for storage in a media having higher latency than the memory. Upon receipt of a request for the content, a cache server computing device transmits the initialization fragment from the memory, retrieves the one or more remaining fragments, and transmits the one or more remaining fragments without retaining the one or more remaining fragments in the memory for subsequent processing.