An innovative approach for the parallel mesh generation has been developed and successfully implemented. The parallelization strategy is based on the domain decomposition concept considered on two levels - the model level and the model entity parametric tree level. The discretization of model entities is based on the generalized parametric tree data structure and application of templates. A very favourable ratio between the computation and communication has been achieved and also satisfactory load balancing has been evidenced. A penalty is paid to outweigh the overall simplicity of the algorithm. The quality of the final mesh depends not only on the mesh size variation but also on the model parameterization. Also the computational performance of the algorithm is dependent on the complexity of the model parameterization. The restrictions on the model topology result in some reduction of modelling flexibility.