Mode Comparison

Understand the differences between masking, generation, and subsetting modes to choose the right approach for your use case.

Overview

TDK supports three primary data transformation modes, each designed for specific use cases. This comparison helps you understand when to use each mode and how they differ in operation.

Mode Comparison Table

Aspect Masking Generation Subsetting

Row count

Preserved

Configurable

Reduced

Source data

Required

Required

Required

Primary keys

Usually preserved

Generated

Preserved

Use case

Anonymize production

Create test data

Extract smaller dataset

Speed

Fast

Fast

Fast (filtering)

Detailed Mode Descriptions

Masking Mode

Purpose: Anonymize sensitive data while preserving structure

  • Replaces sensitive data with realistic fake data

  • Maintains all relationships and constraints

  • Preserves row counts and data structure

  • Best for: Production data anonymization, compliance testing

Learn more about Masking →

Generation Mode

Purpose: Create synthetic data from scratch

  • Generates entirely new data based on schema

  • Configurable row counts and distributions

  • No real production data required

  • Best for: Performance testing, training, demos

Learn more about Generation →

Subsetting Mode

Purpose: Extract a smaller, referentially-intact dataset

  • Reduces dataset size while maintaining relationships

  • Preserves referential integrity across tables

  • Filters data based on specified criteria

  • Best for: Development environments, testing with smaller datasets

Learn more about Subsetting →

Choosing the Right Mode

Quick Decision Guide:

  • Need to anonymize production? → Use Masking

  • Need to create test data? → Use Generation

  • Need a smaller dataset? → Use Subsetting

  • Need anonymization + smaller size? → Combine Masking + Subsetting

Related Pages