This uses a parallel job object (see Section 6.31) to speed up model building. When building multiple models, the optimization process is spread over all nodes in the parallel job (for example, if building 10 models, and you are running on 2 nodes, each node will build 5 models). This feature is still experimental.