Since the last decade the parallel computation has become quite feasible due to the following three aspects. Firstly, a lot of new algorithms, suitable for parallel processing, have been developed. Secondly, the parallel computation ceased to be limited to parallel supercomputers and can be now performed on ordinary computers interconnected by network into a computer cluster. And thirdly, several message passing libraries (typically MPI, PVM), portable to various hardware and operating system platforms, have been developed, which allows to port the parallel applications almost to any platform.
This paper presents an approach for the parallel 3D unstructured mesh generation (Rypl et al. ) that has been implemented on several parallel computing platforms, including IBM SP and PC-based cluster. The results achieved on the last two hardware platforms are presented and compared.