Action Cache Management via Engagement Scores
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Social networking systems face increasing difficulty in providing timely and updated notifications of actions to users as the number of users and connections grows, often resulting in outdated or incomplete notifications due to the need to retrieve data from numerous data stores.
Innovation Solution
A social networking system employs a social data aggregator with an action cache and a cache manager that uses engagement scores and application-specific rules to prioritize and purge data, ensuring that only the most relevant actions are stored in a high-performance memory-resident database for quick access.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If data is retrieved from multiple distributed data stores in real-time, then notification completeness is improved, but system performance and response time deteriorate
Solution Approach 1:
The system performs preliminary actions by caching action data from multiple distributed data stores in advance into a local action cache before notifications are needed. This allows the social data aggregator to have immediate access to action information without querying multiple remote data stores at notification time, thus maintaining complete notifications while improving system performance.
Solution Approach 2:
The patent introduces an intermediary component (action cache) between the distributed data stores and the notification system. This intermediary stores copies of action data locally, allowing the social data aggregator to retrieve information quickly without directly querying multiple remote data stores, thereby resolving the contradiction between completeness and performance.
2Speed
If all action data is cached in memory-resident database, then data access speed is improved, but memory usage and storage costs increase
Solution Approach 1:
The system changes parameters by storing only essential action data fields (action key, user ID, action type, timestamp) in the memory-resident action cache rather than complete action records. This reduces memory usage while maintaining fast access speed for notification generation, as only the most critical parameters are cached in memory.
Solution Approach 2:
The patent applies local quality by storing different types of data in different locations with different access characteristics. Frequently accessed action data is cached in the memory-resident action cache for fast access, while less frequently accessed data remains in remote distributed data stores. This selective caching optimizes both memory usage and data access speed.
3Loss of time
If action cache stores recent actions, then notification timeliness is improved, but data relevance to user engagement deteriorates
Solution Approach 1:
The system implements feedback by using engagement metrics (likes, comments, shares, clicks) to evaluate the relevance of cached action data. The social data aggregator queries the engagement data store to retrieve engagement metrics for actions in the action cache, allowing the system to identify and prioritize highly engaging actions for notifications while maintaining timeliness.
Solution Approach 2:
The patent applies dynamics by making the action cache contents dynamic and adaptive rather than static. The cache is continuously updated with new actions and selectively purged based on engagement metrics and recency. This dynamic management ensures the cache contains both recent actions and highly relevant engaging actions, resolving the contradiction between timeliness and relevance.
Data Source
AI summary
A social data aggregator generates entries of action data describing actions taken by users. A portion of the entries are stored in an action cache to expedite retrieval. To store more recent or relevant entries in the action cache, entries are removed from the action cache based on engagement scores associated with the entries. An engagement score indicates a likelihood of a user requesting content interacting with a notification based on an entry. Entries having the lowest engagement scores or having engagement scores below a threshold are removed from the action cache.