Gradle is releasing a new feature that promises to cut testing time by up to 70%. Predictive Test Selection is an extension to Gradle Enterprise that uses machine learning to select and run tests that are likely to provide useful feedback. In doing so, it avoids running tests that aren’t necessary or insightful.
According to Gradle, Predictive Test Selection can predict over 99% of breaking changes. It also provides explanations for why it avoided certain tests in order to give users more confidence.
It leverages detailed code and dependency data in its predictions, providing insight into production sources and their tests.
In addition, users will be able to try Predictive Test Selection before they deploy it so that they can get a sense of the effectiveness of the model against normal test results.
“Avoiding irrelevant tests which often take minutes or hours has a number of desirable outcomes. For developers, it means less waiting, faster feedback cycles, and a better developer experience,” said Eric Wendelin, lead engineer at Gradle. “For businesses, it means faster time to market without compromising quality, reduced infrastructure costs and energy waste, and happier developers which makes them easier to retain.”
More information about Predictive Test Selection is available here.