The mesh consists of the following mesh entities - nodes, edges, quadrilaterals, and hexahedrons. There are again two basic mesh topologies. In the first one, a mesh entity is described by its nodes - an edge by two, a quadrilateral by four, and a hexahedron by eight nodes. In the second one, a mesh entity is represented by its forming mesh entities of the next lower dimension. Thus an edge is described by two nodes, a quadrilateral by four edges, and a hexahedron by six quadrilaterals. Moreover, in both approaches, each mesh entity is provided with a list of connected mesh entities of the next higher dimension. Thus each node has a list of connected edges, each edge has a list of connected quadrilaterals, and each quadrilateral is associated with up to two hexahedrons. Each mesh entity is also classified to the model entity from which it originates and each model entity, on the other hand, has the list of mesh entities classified to it. The latter topological representation is clearly more advantageous as it offers higher level of topological consistency at smaller memory requirements. In the presented work, however, the former topological representation of the mesh has been employed because a tree-based approach is used for the mesh generation. In this approach, the individual mesh entities (except nodes) arise from templates fitted into the tree, which eliminates the need to store boundary of quadrilateral and hexahedral elements, leading thus to considerable memory savings. Moreover, the tree structure exhibits a high level of topological consistency and therefore only the reduced level of topological consistency is required for the mesh entities. The use of the former topological representation of the mesh is also compatible with the mesh topology employed in the finite element codes. The currently implemented representation of the mesh data structure for the parallel mesh generator is schematically outlined in Table 3.2.