About parallelization of solving of boundary value problems on quasistructured grids

Technological components of boundary problems solution on offered quasi-structured grids of special kind is considered. The feature of these grids is that both macrogrid (coarse grid) in a whole domain and subgrids (local grids) in subdomains are structured and rectangular, it provides efficient structure of data and effective using of computational algorithms. At the same time, resulting quasi-structured grid is adaptive to irregularities within a domain and to complicated shape of domain boundary. It is essential that subgrids can be unmatched. One variant of domain decomposition methods for solving boundary problems is offered, this one is based on separate approximation of boundary problem on the interface and within the subdomains. In order to balance utilization of processors whole set of subdomains is divided into unions (groups) of subdomains. Estimates of paralellization efficiency was obtained for model problem using different number of processors, different grids and different unions of subdomains.