Ad-hoc Graph Definition via Stored Procedures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Relational databases face challenges in efficiently querying and generating dynamic graph structures from input tables, as they require static dependencies and lack the ability to create ad-hoc graph definitions within stored procedures.
Innovation Solution
A method is introduced to generate graph data tables and procedures using table parameters, allowing for dynamic graph object creation and execution of graph algorithms within a relational database, enabling ad-hoc graph definitions and reducing static dependencies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If relational databases use static dependencies for graph structures, then database schema stability is improved, but adaptability for dynamic graph generation deteriorates
Solution Approach 1:
The patent implements dynamic graph generation by allowing graph structures to be created and modified at runtime through stored procedures, rather than requiring static schema definitions. The system generates graph data tables dynamically based on input parameters, enabling the database to adapt to different graph configurations without altering the underlying schema structure.
Solution Approach 2:
The patent uses parameter-driven graph generation where graph structures are defined through input tables and parameters passed to stored procedures. By changing parameters such as vertex tables, edge tables, and graph algorithm specifications, the system can generate different graph structures dynamically while maintaining schema stability.
2Reliability
If relational databases require pre-defined graph structures, then data integrity is improved, but ease of operation for ad-hoc graph definitions deteriorates
Solution Approach 1:
The patent introduces stored procedures as intermediaries between the user and the graph generation process. These procedures encapsulate the complexity of graph structure validation and generation, providing a simplified interface for ad-hoc graph definitions while ensuring data integrity through predefined validation rules and constraints within the procedure logic.
Solution Approach 2:
The patent performs preliminary validation and structure definition within stored procedures before actual graph generation. By pre-defining validation rules, data type constraints, and structural requirements in the procedure code, the system ensures data integrity is maintained while allowing flexible ad-hoc graph definitions through parameter passing.
3Productivity
If relational databases use static graph definitions, then processing efficiency is improved, but adaptability for different graph algorithms deteriorates
Solution Approach 1:
The patent creates a universal stored procedure framework that can handle multiple graph algorithms through parameter specification. The same procedure structure can execute different graph algorithms (e.g., shortest path, connectivity analysis, centrality measures) by accepting different algorithm parameters and input table configurations, maintaining processing efficiency through optimized query plans while achieving versatility.
Data Source
AI summary
A method, a system, and a computer program product for generating graphs using a stored procedure. One or more graph data tables are generated. A procedure having a predetermined name is generated. One or more inputs to the generated procedure includes the generated one or more graph data tables. The generated procedure includes one or more graph objects. The graph data tables are configured to be input to the graph objects. The generated procedure is stored. The stored generated procedure is executed using one or more graph data stored in a graph database.


