Hungry for data ? Check this demo of the synthetic data generation features, in DBJ, to produce realistic data on a relational database.
Either metadata driven, configuration based, or through profiling of existing data set, and supporting constraints like referential integrity, unicity, datatypes domains, you will find the right way to get the synthetic data you need !
The video below presents features I built and provide on "generating synthetic data", with three different methods :
- through database metadata analysis (the datatypes and integrity constraints of tables you shall feed).
- through configuration, by specifying how you want data to be defined (indication on semantics - e.g. this should be an e-mail, this is a temperature measure, these active employee birthdates shall not be before 1950...)
- through data profiling, to build up "real-life like data" (build up data similar to this dataset, that would not be a copy or "scrambling" of it)
These three methods, combined together, can support you to cover a variety of use-cases.
The most evident one is to get data without compromising sensitivity, privacy or confidentiality of "real, production data", to support development and tests environments.
This includes volume stress tests with "massive garbage data synthesis", robustness and resiliency tests with "outliers data synthesis" not available easily otherwise, and user acceptance tests with "realistic datasets synthesis".
You can check the tool details and download it on [ Ссылка ]
[ Ссылка ]
