CFD with adaptive FEM on massively parallel systems

  • Frank Lohmeyer
  • Oliver Vornberger
Part of the Notes on Numerical Fluid Mechanics (NNFM) book series (NONUFM, volume 48)


An explicit finite element scheme based on a two-step Taylor-Galerkin algorithm allows the solution of the Euler and Navier-Stokes equations for a wide variety of flow problems. To obtain useful results for realistic problems, one has to use grids with an extremely high density to obtain a good resolution of the interesting parts of a given flow. Since these details are often limited to small regions of the calculation domain, it is efficient to use unstructured grids to reduce the number of elements and grid points. As such calculations are very time consuming and inherently parallel, the use of multiprocessor systems for this task seems to be a very natural idea. A common approach for parallelization is the division of a given grid, where the problem is the increasing complexity of this task for growing processor numbers. Some general ideas for this kind of parallelization and details of a Parix implementation for Transputer networks are presented. To improve the quality of the calculated solutions, an adaptive grid refinement procedure was included. This extension leads to the need for a dynamic load balancing for the parallel version. An effective strategy for this task is presented and results for up to 1024 processors show the general suitability of this approach for massively parallel systems.


Parallel System Unstructured Grid Division Algorithm Grid Topology Logical Topology 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    A. Vornberger. Strömungsberechnung auf unstrukturierten Netzen mit der Methode der finiten Elemente. Ph.D. Thesis, RWTH Aachen, 1989.Google Scholar
  2. [2]
    W. Koschel and A. Vornberger. Turbomachinery Flow Calculation on Unstructured Grids Using the Finite Element Method. Finite Approximations in Fluid Mechanics II, Notes on Numerical Fluid Mechanics, Vol. 25, pp. 236–248, Aachen, 1989.Google Scholar
  3. [3]
    F. Lohmeyer, O. Vornberger, K. Zeppenfeld and A. Vornberger. Parallel Flow Calculations on Transputers. International Journal of Numerical Methods for Heat and Fluid Flow, Vol. 1, pp. 159–169, 1991.Google Scholar
  4. [4]
    F. Lohmeyer and O. Vornberger. CFD with parallel FEM on Transputer Networks. Flow Simulation with High-Performance Computers I, Notes on Numerical Fluid Mechanics, Vol. 38, pp. 124–137, Vieweg, Braunschweig, 1993Google Scholar
  5. [5]
    F. Lohmeyer and O. Vornberger. Flow Simulation with FEM on Massively Parallel Systems. Computational Fluid Dynamics on Parallel Systems, Notes on Numerical Fluid Mechanics, Vol. 50, pp. 147–156, Vieweg, Braunschweig, 1995.Google Scholar
  6. [6]
    H. Greza, S. Bikker and W. Koschel. Efficient FEM Flow Simulation on Unstructured Adaptive Meshes. In this publication.Google Scholar
  7. [7]
    M. Böhm and E. Speckenmeier. Effiziente Lastausgleichsalgorithmen. Proceedings of TAT-94, Aachen, 1994.Google Scholar

Copyright information

© Friedr. Vieweg & Sohn Verlagsgesellschaft mbH, Braunschweig/Wiesbaden 1996

Authors and Affiliations

  • Frank Lohmeyer
    • 1
  • Oliver Vornberger
    • 1
  1. 1.Department of Mathematics and Computer ScienceUniversity of OsnabrückOsnabrückGermany

Personalised recommendations