Recursive Transition Network for Database Query Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Relational database systems, particularly those based on SQL, face difficulties in formulating complex queries due to grammatically limited degrees of freedom, leading to complicated problem descriptions and inefficient processing, which results in long processing times and reduced user comprehensibility.
Innovation Solution
The method involves forming a recursive transition network (RTN) from independent RTN building blocks with inner directed decision graphs, allowing for arbitrary nesting and recursion, enabling the generation of an optimized access plan that bypasses the structural limitations of traditional query languages like SQL.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If SQL query language with fixed RTN structure is used, then standardization and ease of implementation are improved, but flexibility in query formulation and processing efficiency deteriorate
Solution Approach 1:
The patent segments the RTN into independent, reusable building blocks that can be freely combined and nested. Each building block represents a discrete query processing function, allowing complex queries to be constructed from modular components rather than following a fixed sequential structure.
Solution Approach 2:
The patent implements nested RTN building blocks where smaller RTN structures can be embedded within larger ones, enabling hierarchical query construction. This nesting allows for complex query logic to be built by combining simpler query patterns in a tree-like structure, providing both structure and flexibility.
2Device complexity
If SQL with predetermined processing sequence is used, then system simplicity is improved, but query execution time deteriorates for complex queries
Solution Approach 1:
The patent transforms the static, predetermined processing sequence of traditional SQL RTN into a dynamic structure where the processing order can vary based on the specific query requirements. The system can adaptively determine the optimal execution path through the nested RTN building blocks, allowing for more efficient processing of complex queries.
Solution Approach 2:
The patent performs preliminary analysis and optimization of query execution paths by evaluating multiple possible processing sequences before actual query execution. The system pre-processes the nested RTN structure to identify optimal execution orders, reducing runtime processing time for complex queries.
3Difficulty of detecting and measuring
If fixed RTN decision graph structure is used, then parsing simplicity is improved, but comprehensibility and ease of modification of queries deteriorate
Solution Approach 1:
The patent segments the query structure into distinct, named building blocks that represent logical query components. This segmentation makes queries more readable and understandable, as each building block can be independently modified and reused, improving comprehensibility while maintaining parsing simplicity through structured syntax.
4Reliability
If traditional SQL RTN with limited keyword selection is used, then grammar enforcement is improved, but processing optimization deteriorates
Solution Approach 1:
The patent changes the fundamental parameter of RTN structure from a fixed sequence with limited keyword options to a flexible nested structure with expanded keyword possibilities. This allows the system to enforce grammatical rules through the structured building block syntax while simultaneously enabling optimized processing paths that were previously unavailable in traditional SQL.
Data Source
Figure 1
Figure 2
Figure 3~5
AI summary
The invention relates to a method for controlling a relational database system, wherein a request statement comprising key words is initially analysed using a recursive transition network (RTN). Said RTN consists of RTN modules which are independent from one another and each RTN module has an internal, oriented decision graph which is defined independently from the internal, oriented decision graph of the other RTN modules, said RTN comprising at least one decision position along at least one decision path. In a selection step, the internal decision graphs of all of the RTN modules are transversed by key words, and all possible paths of said decision graphs are transversed until either no coincidence is determined with the respectively selected path in the decision graph and the process is interrupted or the respectively selected path is transversed until the end.