Data Virtualization via Unified Table Name Standardization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data virtualization systems face inefficiencies when accessing tables from multiple data sources with different notation conventions, as they require precise table and column name matching, limiting the ability to access similar data types across disparate sources.
Innovation Solution
A data virtualization apparatus that converts and standardizes table and column names using a thesaurus, rules, or learning models, allowing for the creation of foreign tables with unified names, enabling efficient access and data acquisition from multiple data sources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If table names and column names are defined in different notations across data sources, then each data source can maintain its own naming conventions, but the tables cannot be accessed efficiently through unified queries
Solution Approach 1:
The patent introduces a thesaurus as an intermediary component that maps different table and column names from various data sources to unified standard names. This mediator enables seamless access across heterogeneous data sources without requiring changes to their original naming conventions, resolving the contradiction between maintaining source-specific notations and achieving unified access.
Solution Approach 2:
The system transforms the parameter of table and column names by converting diverse notations into standardized forms through the thesaurus mapping mechanism. This parameter transformation allows the same query to access data from multiple data sources with different original naming conventions, improving both adaptability and operational efficiency.
2Measurement precision
If precise table and column name matching is required, then data accuracy is maintained, but the ability to access similar data types across disparate sources is limited
Solution Approach 1:
The thesaurus serves as a semantic mediator that bridges the gap between precise matching requirements and flexible data access. It maintains accurate data representation by preserving the original table and column names from data sources while simultaneously providing unified standard names for consistent querying across different sources.
Solution Approach 2:
The patent segments the name matching process into two independent layers: the original names layer (preserving data accuracy) and the standardized names layer (enabling flexible access). This segmentation allows the system to maintain measurement precision while achieving adaptability through the thesaurus mapping between the two layers.
Data Source
AI summary
According to one embodiment, a data virtualization apparatus includes a memory and a processor. The processor is configured to acquire first schema information including a first table name of a first source table managed in a first data source, and second schema information including a second table name of a second source table managed in a second data source, convert the first table name into a third table name, and convert the second table name into a third table name, and register first table correspondence information including the first table name and the third table name in the memory, and register second table correspondence information including the second table name and the third table name in the memory.