Direct Server Reply for Edge Infrastructure Services
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In content delivery networks and edge cloud platforms, when a server needs to retrieve content from another server to serve a client, the content often has to traverse the datacenter twice, wasting resources and reducing performance due to unnecessary relay through intermediate servers and occupancy of cache memory.
Innovation Solution
An edge server establishes a transport connection with a client and forwards requests to a cluster server to send content directly to the client, bypassing the need for relay through the edge server, thereby reducing traffic and accelerating delivery by allowing the cluster server to reply directly to the client.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the terminating server obtains content from another server and relays it to the end user, then the content can be delivered to the client, but the content traverses the datacenter twice and consumes valuable processing resources
Solution Approach 1:
The patent extracts the content delivery function from the terminating server and enables the source server to deliver content directly to the end user. This removes the unnecessary relay step where the terminating server would receive and forward content, thereby eliminating wasted processing resources while maintaining delivery capability
Solution Approach 2:
The patent introduces a novel intermediary mechanism - the datacenter network address translation (NAT) device - that enables direct peer-to-peer content transfer between the source server and end user. This intermediary allows the source server to communicate directly with the end user without requiring the terminating server as an intermediate relay point
2Productivity
If the terminating server obtains content from another server, then the content can be served to the end user, but the object occupies valuable cache memory in both servers simultaneously
Solution Approach 1:
The patent extracts the content from the terminating server's cache and delivers it directly to the end user from the source server. This eliminates the duplicate occupancy of cache memory in both servers, as the content only resides in the source server's cache while being delivered directly to the user
3Productivity
If the terminating server relays packets from the source server to the end user, then content delivery is enabled, but valuable processing resources are expended on packet relay
Solution Approach 1:
The patent extracts the packet relay function from the terminating server and transfers it to the source server. The source server now directly communicates with the end user, eliminating the terminating server's workload of receiving and forwarding packets while maintaining the content delivery capability
Data Source
AI summary
An edge server of an infrastructure service establishes a transport connection in user space with a client and in accordance with a transport layer network protocol. The edge server receives a packet over the transport connection with the client that comprises a request for an object. If the edge server cannot serve the object, it forwards the request to a cluster server with an intent indicated for the cluster server to reply directly to the client. The cluster server receives the forwarded request and determines whether to accept the intent indicated by the edge server. If so, the edge server conveys instructions to the cluster server for sending at least a portion of the object directly to the client. The cluster server then sends at least the portion of the object to the client in accordance with the instructions.