Browser Resource Pre-pulling via Idle Network Pre-rendering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional resource pre-pulling methods in browsers are inefficient, resulting in high operational loads and slow page loading speeds due to the need to calculate and select files with high hit ratios for pre-rendering.
Innovation Solution
A method that identifies frequently accessed webpages by obtaining a page address list based on access frequencies and performs pre-rendering when the network is idle, reducing the need for subsequent resource pulling during page loading.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If conventional resource pre-pulling method is used to pre-render pages, then page loading speed can be improved, but the operational load increases and processing time is extended
Solution Approach 1:
The patent applies preliminary action by pre-rendering pages before they are actually needed. The system identifies frequently accessed pages through hit ratio calculation and pre-renders them in advance, so when the user needs to access these pages, they are already ready, thereby improving page loading speed without requiring intensive operational processing at the moment of access.
Solution Approach 2:
The patent implements partial action by selecting only the top N pages with highest hit ratios for pre-rendering, rather than pre-rendering all possible pages. This partial approach focuses computational resources on the most likely to be accessed pages, improving operational efficiency while still achieving significant page loading speed improvements for the critical subset of pages.
2Measurement precision
If hit ratio calculation and file selection is performed for pre-rendering, then relevant resources can be identified, but the operation amount increases and processing time is extended
Solution Approach 1:
The system performs hit ratio calculation and file selection in advance during periods when computational resources are available, rather than performing these operations at the moment of page access. This preliminary identification of relevant resources reduces the processing time required when actually rendering pages, as the resource selection work has already been completed.
Solution Approach 2:
The patent implements continuous monitoring and updating of hit ratios as users access pages, so the system continuously learns user behavior patterns. This continuous action allows the system to maintain accurate resource identification without requiring intensive batch processing, thereby reducing processing time while preserving measurement precision.
Data Source
AI summary
A method and a terminal device for browser resource pre-pulling are provided. A page address list is obtained by processing circuitry of a terminal device that browses webpage via a network. The page address list is obtained from a present webpage in a browser of the terminal device. The page address list includes page addresses of webpages that are linked in the present webpage, and with respective access frequencies satisfying a preset condition. The processing circuitry further determines whether the network is idle. The processing circuitry performs a first page pre-rendering when the network is idle. The first page pre-rendering is to download, via interface circuitry of the terminal device and the network, first page resource of a first webpage according to the first page address, render the first webpage using the first page resource, and store the rendered first webpage in a cache.