Improving feasible interactions among multiple processes

Hierarchical testing requires the verification of individual processes followed by the verification of the interactions among processes. The large number of potential interactions between processes must be managed in order to make the testing process tractable. Fortunately, many potential interactions are actually infeasible and should be ignored during the verification process. Data dependency between processes indicates the potential for a feasible interaction but absolute feasibility can only be determined by evaluating control-flow paths across interacting processes. We propose a method to identify feasible interactions for testing through static analysis combined with the use of a constraint satisfaction programming (CSP) solving engine.