Static Tracker for Web Prefetching Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
HTTP prefetching systems often incorrectly identify or miss child objects, leading to wasted resources and slower user response times due to incorrect prefetching decisions based on flawed models or exceptions.
Innovation Solution
A static tracker system that scans HTTP responses to identify child objects, tracks successes and failures, and adjusts prefetching decisions based on success rates and object characteristics to improve accuracy and resource allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If a prefetcher requests various objects embedded in a webpage in anticipation of those objects being ultimately requested, then the total time required to satisfy the entire series of requests may be reduced, but resources may be wasted to prefetch objects that will never be used
Solution Approach 1:
The system implements feedback mechanisms where the prefetcher monitors which prefetched objects are actually requested by the user and which are not. This feedback information is used to adjust and refine the prefetching model over time, improving accuracy by learning from past successes and failures. The system tracks prefetch hit rates and uses this information to dynamically adjust prefetching decisions.
Solution Approach 2:
The system performs preliminary scanning of webpages to identify embedded objects before making prefetching decisions. By analyzing the webpage structure and content in advance, the system can make more informed decisions about which objects are likely to be requested, reducing both premature and incorrect prefetching actions.
2Extent of automation
If the prefetcher uses models to determine which objects to prefetch, then prefetching decisions can be automated, but incorrect models or difficult to quantify exceptions may lead to repeated incorrect prefetching decisions
Solution Approach 1:
The prefetching system uses dynamic models that adapt and evolve over time based on observed user behavior and prefetch outcomes. Rather than relying on static rules, the system continuously updates its understanding of which objects are likely to be requested, allowing it to handle exceptions and edge cases more effectively while maintaining automated decision-making.
Solution Approach 2:
The system incorporates feedback loops where actual user requests are compared against prefetching predictions. This feedback is used to refine and correct the prefetching model, reducing the impact of initial incorrect models or exceptions. The system learns from its mistakes and improves its reliability over time while maintaining automation.
3Productivity
If the prefetcher scans webpages to identify embedded objects, then more objects can be identified for potential prefetching, but the complexity of determining which objects will be requested increases
Solution Approach 1:
The system segments the webpage into distinct objects and categories (e.g., images, scripts, stylesheets) and applies different analysis rules to each type. This segmentation allows the system to efficiently identify a comprehensive set of potential prefetch candidates while managing complexity by treating different object types with appropriate specialized logic.
Solution Approach 2:
The system employs a universal prefetching framework that can handle multiple types of embedded objects through a common decision-making process. This multi-functional approach allows the system to identify and manage diverse object types without proportionally increasing complexity, as the core prefetching logic remains consistent across different object categories.
Data Source
AI summary
The present invention relates to systems, apparatus, and methods of scanning a response to a first HTTP request for a web page in order to identify a web object for prefetching, and using a static tracker to identify and improve results. In one potential alternative embodiment, after a response is scanned a web object may be prefetched to a proxy server prior to a browser requesting the web object. The proxy server may observe one or more HTTP requests that are associated with the response to the first HTTP request for the web page and measure the success of the prefetching. After success is measured for the specific instance of the web object and the web page, a success rate for prefetching or not prefetching the web object as associated with the web page may be updated.