Failfaster makes your tests fail earlier in a test run.
It does this by comparing the project with the last time it was tested and changing the order the tests run.
Tests are prioritised as follows:
- tests which failed
- tests which are new
- tests which didn't run
- tests which were ok last time
Each category is further sorted by how long the test took to finish last time.
It works by sending a list of commits and the names of all the tests to a server. The server decides the order the tests should run in.