A Grid Reordering Technique for Hybrid Unstructured Flow Solver Based on OpenMP Parallel Environment

Grid reordering is an efficient way to obtain better implicit convergence acceleration in viscous flow simulation based on unstructured hybrid grids, whose data storage is usually random. To improve the matrix quality of the implicit LU-SGS time-stepping scheme and the compatibility of the numerical schemes and OpenMP parallel environment, a grid reordering method for unstructured hybrid grids is proposed. In this method, the structured grids in the viscous layer near-body surface are reordered along the normal direction (like columns) while the unstructured parts are reordered layer by layer in accordance with the neighboring relations. To investigate the performance of the current implementation, 2D and 3D turbulent flows around the RAE2822 airfoil, DLR-F6 wing-body configuration and an aerospace plane are simulated on unstructured hybrid grids. The numerical results show that the grid reordering method is an efficient strategy for improving the convergence rate and the overall efficiency of the flow solver.

[1]  Donald W. Roberts,et al.  Development of a 3-D zonal implicit procedure for hybrid structured-unstructured grids , 1996 .

[2]  A. Jameson,et al.  Lower-upper Symmetric-Gauss-Seidel method for the Euler and Navier-Stokes equations , 1988 .

[3]  Dimitris Drikakis,et al.  Higher-order CFD and interface tracking methods on highly-Parallel MPI and GPU systems , 2011 .

[4]  Yuewen Jiang,et al.  An Improved LU-SGS Implicit Scheme for High Reynolds Number Flow Computations on Hybrid Unstructured Mesh , 2012 .

[5]  Kazuhiro Nakahashi,et al.  Applications of unstructured hybrid grid method to high‐Reynolds number viscous flows , 1999 .

[6]  Wang Gang,et al.  Reordering of 3-D Unstructured Grids for Computing Efficiency Improvement , 2005 .

[7]  Yuanli Wang,et al.  Hybrid Mesh Generation for Viscous Flow Simulation , 2006, IMR.

[8]  Ye Zheng.  Generation of Three Dimensional Mixed and Unstructured Grids and its Application in Solving Navier-Stokes Equations , 2003 .

[9]  Kazuhiro Nakahashi,et al.  Reordering of 3-D hybrid unstructured grids for vectorized LU-SGS Navier-Stokes computations , 1997 .

[10]  Seokkwan Yoon,et al.  An LU-SSOR scheme for the Euler and Navier-Stokes equations , 1987 .

[11]  Victor E. Malyshkin,et al.  Parallel computing technologies , 2011, The Journal of Supercomputing.

[12]  Oleg Bessonov.  OpenMP Parallelization of a CFD Code for Multicore Computers: Analysis and Comparison , 2011, PaCT.

[13]  Vincent Moureau,et al.  Design of a massively parallel CFD code for complex geometries , 2011 .

[14]  Jay Hoeflinger,et al.  Producing scalable performance with OpenMP: Experiments with two CFD applications , 2001, Parallel Comput..

[15]  William Gropp,et al.  High-performance parallel implicit CFD , 2001, Parallel Comput..