Unified Data Model for Flexible Database Schema
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional business information (BI) systems require extensive modifications to their database schema and associated components whenever the data model needs to be updated, leading to high costs and long project timelines due to their hardwired nature, making it difficult to extend or modify these systems efficiently.
Innovation Solution
A flexible database schema is implemented using a unified data model (UDM) with an extract, transform, load (ETL) module that transforms and loads data into key-value pair tables, allowing for dynamic representation of entities, relationships, and results without requiring structural changes, enabling the inclusion of new data types without altering the schema.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional BI systems use explicit database schema structures, then database performance is optimized for the current data model, but any updates to the data model require extensive modifications to the schema and all associated components
Solution Approach 1:
The patent implements a dynamic database schema where the data model can be updated without requiring extensive modifications to the schema structure or associated components. The system allows for flexible changes to entity relationships and data types while maintaining performance through the key-value pair storage mechanism, resolving the contradiction between optimized performance and model flexibility.
Solution Approach 2:
The patent enables parameter changes in the data model without structural modifications to the database schema. By using a unified data model that stores data as key-value pairs with metadata, the system can adapt to new data types and relationships by changing parameters rather than restructuring the entire schema, thus maintaining both performance and flexibility.
2Stability of the object's composition
If traditional BI systems use hardwired data structures, then the system structure is stable and well-defined, but the cost of extending or modifying the system is prohibitive and project timelines are extended
Solution Approach 1:
The patent introduces dynamic capabilities to the data model while maintaining structural stability through a unified schema-based approach. The system allows for rapid updates to data models without requiring changes to the underlying schema or associated components, thereby reducing project timelines while maintaining stability.
Solution Approach 2:
The patent segments the data model into distinct entities, relationships, and attributes that can be independently modified. This segmentation allows for targeted changes to specific parts of the data model without affecting the entire system structure, reducing the time and cost associated with modifications.
3Reliability
If traditional BI systems require modifications to ETL processes and queries when data model changes are introduced, then data integrity is maintained, but the complexity of system modifications increases
Solution Approach 1:
The patent introduces a unified data model as an intermediary layer between the database schema and the ETL processes. This intermediary allows for data model changes without requiring modifications to ETL processes or queries, as the unified model abstracts the underlying schema changes, thereby maintaining data integrity while reducing modification complexity.
Solution Approach 2:
The patent uses a copy of the data model in the unified schema that can be independently updated. This copy allows for changes to be made without affecting the original ETL processes and queries, as they continue to operate against the unified model structure, thereby maintaining integrity while reducing complexity.
Data Source
AI summary
Various systems and methods can provide a flexible database schema. One method can store information identifying a first entity in a first table. A unified data model includes several tables, including the first table and a metadata table. Each of the tables stores information describing one or more entities belonging to a respective archetype. The first table already stores information identifying a second entity when the information identifying the first entity is stored. The second entity is already related to another entity when the information identifying the first entity is stored. The first entity is a new type of entity not already stored in the first table when the information identifying the first entity is stored. The method then stores metadata associated with the first entity in a metadata table. The metadata then identifies the new type of entity.