Synthetic Data Generation for Database Query Testing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing database management systems (DBMS) face challenges in testing without resulting queries due to database unavailability, sensitive information constraints, and the need for benchmarking quality levels through executing database queries.
Innovation Solution
A data generation module generates data records to satisfy database queries, enabling testing by updating the database content and inspecting DBMS performance, with a query obtainer obtaining and processing database queries to generate records that modify the database state, allowing for effective benchmarking and quality assurance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If database queries are executed to test DBMS performance, then quality assurance and benchmarking are enabled, but database unavailability or sensitive information constraints prevent testing
Solution Approach 1:
The system performs preliminary actions by generating test data records in advance before executing database queries. The data generation module creates synthetic data that satisfies the requirements of test queries, ensuring the database is properly populated and ready for testing without requiring access to existing sensitive databases.
Solution Approach 2:
The patent introduces an intermediary data generation module that acts as a mediator between the testing system and the database. This module generates synthetic test data that can be inserted into the database, enabling queries to be executed without direct access to sensitive existing data, thus overcoming database unavailability or sensitivity constraints.
2Adaptability or versatility
If existing databases are used for testing, then real-world data scenarios are available, but sensitive information constraints and database unavailability prevent access
Solution Approach 1:
The system creates copies of real-world data scenarios through synthetic data generation. The data generation module produces test records that replicate the structure and characteristics of real database data without using actual sensitive information, thereby maintaining testing scenario diversity while avoiding data loss from inaccessible databases.
Solution Approach 2:
The patent applies parameter changes by transforming real database schemas into synthetic data structures. The system modifies data parameters and characteristics to create test records that mirror real-world scenarios while using anonymized or generated values, enabling diverse testing without accessing sensitive existing data.
3Ease of operation
If data records are generated to satisfy queries, then testing can proceed without existing database content, but data generation complexity increases
Solution Approach 1:
The data generation module is designed with universality to handle multiple query types and database schemas through a single integrated system. It provides multi-functional capabilities including schema analysis, synthetic data generation, and query execution, thereby improving testing accessibility while managing complexity through consolidation rather than multiplication of components.
Data Source
AI summary
A method and apparatus for generating at least one data record in respect to a database query comprising a fetch command. A database may be updated according to the at least one data record. In an exemplary embodiment, a database management system may be tested by performing the database query against the database. In another exemplary embodiment, a data record that satisfies the database query is generated in order to increase coverage when testing a database management system.