Acceleration Server Caching Multi-Domain Web Content
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In a multi-domain web page configuration, clients face limitations in simultaneously acquiring content data due to restricted TCP sessions and the need for server push distribution, which complicates the distribution of web content across multiple web servers.
Innovation Solution
A distribution device that caches the main HTML file and sub-content data from multiple domains on an acceleration server, allowing for server push distribution and rewriting link destinations and file names to facilitate efficient content delivery.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If content data is distributed across multiple web servers with different domains, then content can be arranged in a distributed manner, but the client cannot simultaneously acquire all content data due to limited TCP sessions
Solution Approach 1:
The patent combines multiple content data from different domains into a single response transmitted over one TCP connection. The acceleration server aggregates the HTML file and multiple content data (images, CSS, JavaScript, etc.) from different web servers and delivers them together to the client through a unified communication channel, eliminating the need for separate TCP sessions for each content acquisition.
Solution Approach 2:
The acceleration server acts as an intermediary between the client and multiple web servers. It receives a single request from the client, automatically retrieves necessary content data from multiple distributed web servers, and delivers the aggregated content in one response. This intermediary approach allows the client to interact with only one server while the acceleration server handles the complexity of multi-domain content aggregation.
2Productivity
If the client accesses multiple web servers to acquire content data, then all content can be acquired, but the distribution process cannot proceed until the HTML file acquisition is completed
Solution Approach 1:
The acceleration server performs preliminary actions by pre-fetching and caching content data from multiple web servers before the client requests it. When the client requests a web page, the acceleration server has already retrieved and stored the HTML file and associated content data (images, CSS, JavaScript) in its buffer, allowing immediate delivery without sequential access delays.
Solution Approach 2:
The acceleration server maintains continuous operation by parallelizing the acquisition of HTML files and content data. Instead of waiting for sequential completion, it continuously retrieves multiple content items simultaneously from different web servers, keeping the data flow continuous and eliminating idle waiting time in the distribution process.
3Extent of automation
If server push distribution is implemented, then content can be distribured initiatively, but it is not compliant with existing TCP protocols
Solution Approach 1:
Instead of implementing server push where the server initiates data transmission, the patent inverts the approach by using client pull with automatic aggregation. The client requests content, and the acceleration server automatically responds with aggregated content data that would otherwise require multiple separate requests. This inversion achieves efficient content delivery while maintaining compatibility with standard client-initiated HTTP/TCP protocols.
Data Source
AI summary
The present invention provides a distribution device for web content capable of quickly distributing a multi-domain web page in response to a request from a client. A unit 1 acquires from a web server an html file A and related sub content B, C, D for a web page requested from a client, and then caches the html file A and related sub content B, C, D in a unit 2. A unit 3 edits each piece of content data B, C, D so that the client can obtain the html file A and each piece of content for the web page from the same domain. A unit 701 reads out the html file A that was cached in the unit 2 and distributes the html file A to the client. A unit 703 push-distributes each piece of content data B, C, D to the client.