#BuildPropulsionLab at #DevoxxUK: @jbaruch with @inponomarev
summary
Synthesized product generates synthetic relational data from schema, metadata or small samples and matches statistical characteristics (for example prices concentrated between $50–$150, realistic-looking credit card formats, names and addresses) so tests run against representative values without leaking secrets. It supports masking, subsetting and full-from-scratch generation and exposes a Testcontainers plugin on Maven Central that seeds containers with non-empty, realistic datasets and synthesizes data matching statistical distributions to speed integration tests and reproduce edge behaviours.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 0:48Data synthesis importance
- 2:54Synthetic realistic test data
- 5:42Developer productivity challenges
- 6:23Mocks vs test containers
- 7:26Test containers limitations
- 8:51Mocks outperform real systems
- 10:17Production testing data volume
- 12:24Mocks and containers limits
- 13:08Synthetic data for test containers
- 14:15Integration testing with real data
related talks
30m🤯 No data? No problem! Synthetic data to the rescue
47mEfficiency up, failed releases down; compliant production data for your test and dev environments
43mAccelerate Release Velocity and Quality with Modern DevOps Test Data Management Practices
19mDevDataOps - Data-Driven Product Development for Software Engineers
40mSynthetic data: when, why, and how
39m"Unit tests for Data" with Deequ
38mEnabling Big Data and Machine Learning for the Masses...
20mData Mesh testing: An opinionated view of what good looks like