| The primary objective of this study is to develop an efficient, scalable, parallel incompressible flow solver capable of performing viscous, high Reynolds number flow simulations for complex geometries using multielement unstructured grids. The present parallel unstructured viscous flow solver is based on domain decomposition for concurrent solution within subdomains assigned to multiple processors. The solution algorithm employs iterative solution of the implicit approximation, and its software implementation uses MPI message passing for interprocessor communication.; Key parallelization issues addressed in this work are (1) definition of the iteration hierarchy, (2) treatment of connectivity between subdomain interfaces, and (3) methods for coupling of subdomains. A heuristic, semiempirical performance estimate is developed and evaluated. With this performance estimate, scalability characteristics of the solution algorithm may be calculated for a particular architecture and/or predicted for a given problem a priori.; Validation and verification of the solution procedure are carried out on several small steady and unsteady model problems with excellent agreement to experimental, theoretical, and numerical results. The present parallel flow solver is demonstrated for large-scale meshes with viscous sublayer resolution (y+ ∼ 1) and approximately 106 points or more. Complex geometry 3D applications include (1) a full-scale ship hull, (2) a SUBOFF model hull with stern appendages, (3) a fully-configured high-lift transport, and (4) a maneuvering tiltrotor aircraft. The first three computations are shown to agree well with available experimental data. The maneuvering tiltrotor aircraft simulation is a demonstration of capability for the parallel solution algorithm in the context of an extremely complex geometry and unsteady flowfield. |
