Knowledge Graph Data Fragmentation with Adaptive Diffusion Balancing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data fragmentation methods for large-scale knowledge graphs fail to balance the distribution of fragmented data across multiple devices, leading to inefficiencies in storage and query processes.
Innovation Solution
A method and apparatus for data fragmentation on knowledge graphs that adjust diffusion velocities based on comparisons between fragmented nodes and edges across devices to achieve balanced distribution, using a first diffusion velocity to select and distribute edges, and adjust this velocity based on node and edge balance degrees to optimize allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Volume of stationary object
If data fragmentation is performed on large-scale knowledge graphs using distributed storage, then the knowledge graph can be stored across multiple devices, but the distribution of fragmented data becomes unbalanced across devices
Solution Approach 1:
The patent implements dynamic adjustment of diffusion velocity during the fragmentation process. The system monitors the balance degree of fragmented nodes and edges across devices in real-time, and adjusts the diffusion velocity parameter dynamically to maintain balanced distribution. This transforms a static fragmentation process into a dynamic one that adapts to changing distribution states.
Solution Approach 2:
The patent employs feedback mechanisms where the system continuously monitors the distribution state of fragmented data across devices. Based on the monitored balance degree, the system provides feedback to adjust the diffusion velocity, creating a closed-loop control system that ensures balanced data distribution while maintaining storage scalability.
2Productivity
If diffusion velocity is increased to speed up data fragmentation, then fragmentation process completes faster, but data distribution becomes more unbalanced
Solution Approach 1:
The patent makes the diffusion velocity dynamic rather than fixed. The system adjusts diffusion velocity based on real-time monitoring of data distribution balance across devices. When balance deteriorates, velocity is reduced; when balance is maintained, velocity can be increased, optimizing both speed and distribution quality.
Solution Approach 2:
The patent changes the diffusion velocity parameter adaptively during the fragmentation process. By modifying this key parameter based on distribution balance feedback, the system optimizes the trade-off between fragmentation speed and distribution balance, preventing the harmful effects of both too-high and too-low velocity settings.
3Manufacturing precision
If diffusion velocity is decreased to improve data distribution balance, then fragmentation process becomes slower, but storage efficiency is improved
Solution Approach 1:
The system dynamically adjusts diffusion velocity rather than using a fixed low value. This allows the system to achieve good distribution balance only when necessary, while maintaining higher speeds when distribution is already favorable, thus optimizing the speed-balance trade-off.
Solution Approach 2:
The patent implements adaptive parameter changes where diffusion velocity is modified based on distribution balance requirements. This selective parameter adjustment ensures that low velocity (and reduced speed) is applied only when needed to maintain balance, rather than being applied continuously.
4Volume of stationary object
If initial splitting is performed on all edges, then data can be distributed to multiple devices, but control over the fragmentation process is lost
Solution Approach 1:
The patent implements feedback control where the system monitors the state of data distribution across devices and uses this information to adjust the fragmentation process. This feedback mechanism restores controllability by enabling real-time adjustments based on actual distribution outcomes.
Solution Approach 2:
The system transitions from a static initial splitting approach to a dynamic controlled fragmentation process. By continuously monitoring distribution state and adjusting diffusion velocity dynamically, the system regains control over the fragmentation process while maintaining distributed storage benefits.
Data Source
AI summary
Embodiments of this specification provide methods and apparatuses for performing data fragmentation on a knowledge graph. The method is used to split a knowledge graph into a plurality of pieces of data respectively included in a plurality of devices. First, initial splitting is performed on a plurality of edges in the knowledge graph. Any first device selects a diffusion node from end nodes of a first part of edges; obtains, as a to-be-fragmented edge, an edge that is in the knowledge graph and that uses the diffusion node as an end node on one side; and adds a target edge in the to-be-fragmented edge to first fragmented data of the first device. Then, the first device obtains a fragmented node in fragmented data of another device; adjusts the first diffusion velocity; and continues to select a diffusion node based on an adjusted first diffusion velocity, and cyclically performs the step.