Synthetic data (Preview)

Delphix Synthetic Data enables teams to generate realistic, production-like test data that does not exist in production. Unlike masking, which transforms existing production data, Synthetic Data creates entirely new datasets from scratch. This allows teams to support testing scenarios that production data alone cannot cover.

This preview release introduces end-to-end synthetic data generation fully managed within DCT. Teams can connect to a reference source, allow the system to analyze the schema, generate synthetic data, and deliver it to one or more target environments without deploying separate nodes or engines.

How this helps

Development and testing teams often face a key limitation: they can only test with existing production data. This creates gaps in scenarios such as:

• Building new applications without production history (greenfield development)

• Testing schema changes or new features before production data reflects them

• Validating edge cases, negative scenarios, or error handling not present in production

Delphix Synthetic Data removes these limitations. It generates structurally valid and referentially consistent datasets across schemas and systems, enabling teams to test freely without relying on production data.

What's included in this preview

This preview delivers the first end-to-end synthetic data workflow fully configurable and managed within DCT. The following building blocks are now available:

• Reference Connectors: Connect to source databases for schema context

• Target Connectors: Define one or more environments for data delivery

• Generators: Configure how synthetic values are created per table and column

• Synthetic Data Jobs: Define end-to-end workflows that combine rules and targets

Core features

Schema-aware generation

The system analyzes the reference schema, including table structures, data types, constraints, and relationships. It uses this understanding to generate data that remains structurally valid and consistent across tables.

Multi-target delivery

A single schema analysis supports data generation across multiple non-production environments.

Fully embedded in DCT

All configuration, execution, and monitoring run within DCT, eliminating the need for external infrastructure.

Key capabilities

Capability	Details
Database support	Oracle. Additional databases planned for future releases.
Delivery targets	One or multiple targets supported.
Schema analysis	Automatic analysis of reference source schema to inform generation.
Infrastructure	Fully embedded in DCT.
Enriching existing datasets	This release only supports greenfield scenarios. Enriching existing datasets with synthetic data will be supported in upcoming releases.

Process workflow

1. Enable the Synthetic Data feature flag by contacting your account team or emailing delphix-early-adopters@perforce.com.

2. Connect to a reference Oracle database whose schema will be used as the basis for generation.

3. Delphix Synthetic Data analyses the reference schema, examining table structures, column data types, constraints, relationships and semantics.

4. Configure generation rules to define how synthetic values are produced for each table and column.

5. Define one or more target connectors - the environments where synthetic data will be delivered.

6. Run the synthetic data job to generate data into configured targets.

Architecture highlights

Single control plane

Manage reference configuration, schema analysis, generator setup, execution, and monitoring centrally within DCT.

Reference-driven generation

DCT connects to the reference source over a read-only JDBC connection and queries only system catalog metadata (table definitions, column types, constraints, indexes, and relationships). No table data is read, extracted, or cached. As data generation is driven entirely by schema structure, the reference source can be any representative environment — it does not need to be production.

Multi-target by design

The architecture supports fan-out delivery natively. Use a single configuration to generate and distribute synthetic data across multiple environments.

Feedback

Feedback during the preview stage is essential to shaping the general availability release. If you encounter issues, have questions, or want to share suggestions about Delphix Synthetic Data, contact the team at delphix-early-adopters@perforce.com.