Layered Graph Data Structure for Efficient Querying
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional relational database systems are inefficient for storing graphical data models, leading to increased resource consumption, slow query response times, and high storage costs due to their inherent differences from graphical data structures, particularly in domains like social networks and product catalogs.
Innovation Solution
A layered graph data structure is introduced, utilizing a vertex table and an edge table to store data, with sharding techniques to distribute storage across multiple servers, optimizing storage and query performance by aligning the database structure with the graph data domain.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If relational database systems are used to store graphical data models, then data can be stored in structured tables, but storage requirements increase and query response times slow down
Solution Approach 1:
The patent segments the storage of graphical data models by creating separate table structures for different components of the graph (vertices table, edges table, and relationship tables). This segmentation allows the database to store and query different parts of the graph structure independently, improving query response times while maintaining reliable data storage.
Solution Approach 2:
The patent introduces a new dimension to relational database storage by implementing a multi-layered table structure that represents graph hierarchical levels. This dimensional transformation allows the database to efficiently store and query graph data across multiple levels of abstraction, resolving the contradiction between storage reliability and query speed.
2Reliability
If traditional relational database structures are used for graphical data, then data can be organized in tables, but storage efficiency decreases and resources are consumed unnecessarily
Solution Approach 1:
The patent applies local quality by creating specialized table structures tailored to specific graph components. Instead of using a generic relational structure, the database implements vertex tables for node data, edge tables for relationship data, and relationship tables for connectivity information. This localized optimization reduces storage requirements while maintaining reliable data organization.
Solution Approach 2:
The patent changes the structural parameters of the database by transforming traditional flat table structures into hierarchical multi-layered tables that reflect the graph data model. This parameter transformation optimizes storage efficiency by organizing data according to graph theory principles rather than conventional relational structures.
3Adaptability or versatility
If the number of tables increases to model complex relationships, then data can be stored comprehensively, but search time increases exponentially
Solution Approach 1:
The patent applies preliminary action by pre-organizing graph data into hierarchical layers during the data loading phase. Relationship tables and edge tables are structured in advance to reflect the graph's connectivity patterns, allowing the database to perform efficient searches without requiring complex joins across multiple tables during query execution.
Solution Approach 2:
The patent resolves the search time issue by introducing a hierarchical layer dimension to the table structure. Instead of using numerous flat tables, the database organizes data across multiple hierarchical levels, allowing queries to traverse the hierarchy efficiently rather than performing exponential searches across many tables.
Data Source
AI summary
A layered graph data structure can be stored using a vertex table and an edge table. The vertex table includes a vertex identifier column and a first graph identifier column. The edge table includes a second graph identifier column, a node identifier column, and an edge type column. Queries of the layered graph data structure include a target entity and a graph level, and iterative searching of the vertex table and the edge table is performed based on the graph level and data stored in edge type column.