In general, techniques are described for reducing response times to retrieve content in an intermediate network device. In particular, the intermediate network device receives a packet from a client device of a first network that requests content from a remote network device of a second network, inspects the packet to determine whether the requested content has been previously cached to either of a first and a second memory of the device, issues a request to load the requested content from the second memory to the first memory based on the determination and queues the packet within in the queue. After queuing the packet, the intermediate network device then processes the packet to assemble a response that includes the content from the memory.