Abstract:
The verification of temporal properties against a given system may require the exploration of its full state space. In explicit model-checking this exploration uses a Depth-First-Search (DFS) and can be achieved with multiple randomized threads to increase performance. Nonetheless the topology of the state-space and the exploration order can cap the speedup up to a certain number of threads. This paper proposes a new technique that aims to tackle this limitation by generating artificial initial states, using genetic algorithms. Threads are then launched from these states and thus explore different parts of the state space. Our prototype implementation runs 10% faster than state-of-the-art algorithms. These results demonstrate that this novel approach worth to be considered as a way to overcome existing limitations.