Test Data Generation

Generation Process

Overview

Describes the end-to-end workflow for generating test data from multi-table metadata.

Process Steps

1

Table generation order

Based on the foreign key relationships, an order for table data generation is determined.

2

Record count calculation

The number of records to generate for each table is determined based on maximum subject references. If no references are present, the original table size ratios are used.

3

Data generation

Using the bounds, categories, semantic types, and other details provided in the metadata, the test data is generated. Key constraints are also taken care of during the generation process.

4

Format conversion and upload

Based on the chosen output format, the generated data is written to files and uploaded to a pre-configured S3 bucket.