DataGeneration
DataGeneration
Overview & Challenge
Conventional Test Data Management (TDM) uses Production as the primary source of reference data which is loaded into test environments to support testing. While Production data is a logical starting point to source representative data for testing purposes, it does not cater to all TDM requirements, and it carries substantial data security risk. To add to this challenge, GDPR regulations in Europe now forbid the use of Production data as a source for testing activities. Production data limitations include:
- Production data does not contain full or even optimal coverage (usually < than 50%)
- Data for new build cannot be sourced from Production
- Using Production as the primary source of data poses a customer data security risk
Solution
Production is a logical place to start, especially when it comes to capturing an understanding of your data landscape and the relationships that need to be maintained for referential integrity, but at the very least it needs to be augmented with the generation of synthetic data on demand. Generated data is required to:
- Fill the gaps in Production coverage
- Support new build
- Support virtual services & stubbing frameworks
Longer term, especially once the cross-system data model has been accurately captured, data generation should take on more and more of your data provisioning requirements.
How does Orson Work
Orson’s Data Generation module is designed to generate data on demand to support these requirements but it does much more than that. Orson’s DG module:
- Generates ‘environment data’ based on calculated optimized coverage
- Calculates expected results for each input variation for a given business process
- Matches the right data to the right tests – automatically, based on selection rules
- Does all of this ‘in bulk’ instead of 1 by 1, dramatically reducing TDM time
Orson then feeds the right data on demand to test execution tools and virtual services.
Benefits
Orson’s ability to generate data on demand, based on optimal coverage, automatically calculate expected results and assign data to specific tests based on ‘test coverage criteria’ is transformational, delivering the following benefits:
- Eliminates data security risk with synthetic data
- Test optimal coverage – generating the data you don’t have otherwise
- Button-press creation of synthetic data, based on optimal coverage, enables Continuous Testing and increases development velocity
- Generate data on demand to fit robust testing in-sprint
- Calculate expected results to reduce human error and business validation overhead
- Match the right data to 1,000s of tests automatically – instead of 1 by 1, manually
- Store the intelligence that underpins all of this as a set of reusable rules – minimizing SME bottlenecks and persisting ‘coverage and data intelligence’
