Design and implementation of an automatic parallelization of state space generation

It's hard for a user ignorant of parallel programming to resolve models by numerical analysis method in distributed environments, which normally involves state space generation (SSG) in parallel. To lower this threshold, an automatic parallelization approach based on MapReduce framework is presented in this paper. It has been implemented in a small-scale distributed environment, and its correctness and feasibility has been verified by the results of experiments carried on a series of models varying in scale.