It is worth detailing some major features that we expect from a Test Data Management (TDM) product. Firstly, it should be able to deal with sensitive data. That is, either the product needs to include data masking capabilities or it needs to be able to generate synthetic data, or both. Secondly, the solution should implement the principles of DevOps and support agile environments. This means putting as little strain as possible on database administrators and, in practice, this either means generating virtual copies of your source database(s) or running off a test data warehouse. In either case, the idea is to allow the re-provisioning of (amended) test data on demand, without having to go back to the production data or the administrators thereof. In effect, providing a self-service test data environment for developers and testers. Thirdly, you would like your TDM solution to integrate with other tools within the testing environment so that, for example, test data can be automatically provisioned to relevant test cases. Lastly, you would like your TDM solution to be not just representative of your real data but also to include outliers, boundary conditions and erroneous data that might not – almost certainly will not – appear in your production data. This Market Update compares TDM products on this basis.
This paper has been superseded by Test Data Management (2021) (July, 2021)