In order to check the correctness of the integration process of different system parts, integration testing is performed. This type of checking is focused on functional testing, checking reliability and performance of the integrated system components of the developed product.
Firstly, unit tests are executed separately for each system component and only then their integration process checking is performed - whether the combination of these modules gives an expected result.
Such type of checking may be conducted by both developers and testers. During the fulfillment of the integration tests, one may apply three approaches.
What are integration testing approaches?
- “Big bang” approach
- “Top-down” method
- “Bottom-up” method
According to the “big bang” method, all modules/units are integrated and checked simultaneously. Mainly, such checking type is executed when the system is ready for testing integration as an entity. But this is not system testing as only integrated units are checked, not the whole product system.
Such approach helps to save time but if a severe error is detected, then it may be hard to define its reason.
The “top-down” testing method is executed gradually, step-by-step. In order to check first unit test, the specialists use stubs and only then the modules are integrated. But in this way, the main functional is checked at the end.
Performing bottom-up testing, the specialists use certain program stimulators - drivers. Modules/units checking is performed bottom-up gradually till all unit data levels will be integrated and tested as an entire system.