RiTMO: A Method for Runtime Testability Measurement and Optimisation

Runtime testing is emerging as the solution for the integration and assessment of highly dynamic, high availability software systems where traditional development-time integration testing is too costly, or cannot be performed. However, in many situations, an extra effort will have to be invested in implementing appropriate measures to enable runtime tests to be performed without affecting the running system or its environment. This paper introduces a method for the improvement of the runtime testability of a system, which provides an optimal implementation plan for the application of measures to avoid the runtime tests' interferences. This plan is calculated considering the trade-off between testability and implementation cost. The computation of the implementation plan is driven by an estimation of runtime testability, and based on a model of the system. Runtime testability is estimated independently of the test cases and focused exclusively on the architecture of the system at runtime.