Progressive Prefetching With HTTP Object Dependency Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing web page prefetching systems often waste resources by incorrectly prefetching objects that are not needed, leading to slower performance and increased wait times due to missed objects and difficulty in determining dependencies between HTTP objects.
Innovation Solution
A system that analyzes user browsing indicators and identifies interdependencies between HTTP objects, delaying prefetching of secondary objects until primary objects are fetched, and optimizing caching priorities based on statistical correlations and user usage data to improve prefetching efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If prefetching is performed for all embedded objects in a web page, then the likelihood of satisfying future requests locally increases, but resources are wasted prefetching objects that are not later requested by the user
Solution Approach 1:
The system performs preliminary actions by prefetching objects before they are actually requested, but uses a predictive model to determine which objects are likely to be requested. This allows the system to advance the timing of useful prefetch operations while avoiding waste on objects that won't be used.
Solution Approach 2:
The system changes the parameter of prefetch decision-making from a binary approach (prefetch or not) to a probabilistic approach based on predictive modeling. By calculating likelihood scores and using statistical correlations from usage data, the system dynamically adjusts which objects to prefetch based on predicted user behavior patterns.
2Reliability
If the prefetcher requests various embedded objects in anticipation of user requests, then more objects may be available locally, but it becomes difficult to determine which objects will ultimately be requested
Solution Approach 1:
The system implements feedback mechanisms by collecting actual user usage data and comparing it with prefetch predictions. This feedback is used to continuously refine and update the predictive models, improving accuracy over time while managing complexity through iterative learning from real user behavior patterns.
Solution Approach 2:
The system introduces an intermediary predictive modeling layer between the prefetcher and the objects. This intermediary uses statistical correlations and usage data to translate complex determination of which objects to prefetch into manageable probability assessments, simplifying the decision-making process.
3Loss of energy
If incorrect objects are prefetched repeatedly based on incorrect models, then the prefetching system may waste resources, but correcting the model requires identifying and analyzing exceptions to rules
Solution Approach 1:
The system uses feedback from actual user requests to detect when prefetch predictions are incorrect. By monitoring whether prefetched objects are actually requested, the system identifies model errors and uses this information to correct and refine predictive rules, reducing resource waste while systematically improving accuracy.
Data Source
AI summary
The present invention relates to systems, apparatus, and methods of using usage data to determine the dependency structures of a web application, including dependency structures between follow-on objects of an initial object in a web transaction. In one embodiment, an input URL and associated dynamic response data are analyzed for such nested or dependent relationships. In further embodiments, analysis of these relationships are used to improve prefetching operations to lower overall page load times.