Implementation of implicit finite element methods for incompressible flows on the CM-5. (English) Zbl 0848.76037

Kennedy, J. G.; Behr, M.; Kalro, V.; Tezduyar, T. E.

Comput. Methods Appl. Mech. Eng. 119, No. 1-2, 95-111 (1994).

A parallel implementation of an implicit finite element formulation for incompressible fluids on a distributed-memory massively parallel computer is presented. The dominant issue that distinguishes the implementation of finite element problems on distributed-memory computers from that on traditional shared-memory scalar or vector computers is the distribution of data (and hence workload) to the processors and the non-uniform memory hierarchy associated with the processors, particularly the non-uniform costs associated with on-processor and off-processor memory references.
The discussion relies primarily on abstract features of the hardware and software environment and should be applicable, in principle, to a variety of distributed-memory system. The actual implementation is carried out on a Connection Machine CM-5 system with high performance communication functions.

MSC:

76M10	Finite element methods applied to problems in fluid mechanics
76D05	Navier-Stokes equations for incompressible viscous fluids
65Y05	Parallel numerical computation

Keywords:

distributed-memory massively parallel computer; non-uniform memory hierarchy

References:

[1]	Malone, J. G., Automated mesh decomposition and concurrent finite element analysis for hypercube multiprocessor computers, Comput. Methods Appl. Mech. Engrg., 70, 27-58 (1988) · Zbl 0634.73062
[2]	Farhat, C.; Wilson, E., A new finite element concurrent computer program architecture, Internat. J. Numer. Methods Engrg., 24, 1771-1792 (1987) · Zbl 0625.73080
[3]	Lyzenga, G. A.; Raefsky, A.; Hager, B. H., Finite elements and the method of conjugate gradients on concurrent processors, (Caltech/JPL Report C3P-119 (1984), California Institute of Technology: California Institute of Technology Pasadena, CA)
[4]	Mathur, K. K.; Johnsson, S. L., The finite element method on a data parallel computing system, Internat. J. High-Speed Comput., 1, 29-44 (1989) · Zbl 0725.73088
[5]	Belytschko, T.; Plaskacz, E. J.; Kennedy, J. M.; Greenwell, D. L., Finite element analysis on the Connection Machine, Comput. Methods Appl. Mech. Engrg., 81, 229-254 (1990) · Zbl 0728.73068
[6]	Shapiro, R. A., Implementation of an Euler/Navier-Stokes finite element algorithm on the Connection Machine, (AIAA 29th Aerospace Sciences Meeting. AIAA 29th Aerospace Sciences Meeting, AIAA-91-0438 (1991))
[7]	Farhat, C.; Sobh, N.; Park, K. C., Transient finite element computations on 65536 processors: The Connection Machine, Internat. J. Numer. Methods Engrg., 30, 27-55 (1990)
[8]	Johan, Z.; Hughes, T. J.R.; Mathur, K. K.; Johnsson, S. L., A data parallel finite element method for computational fluid dynamics on the Connection Machine system, Comput. Methods Appl. Mech. Engrg., 99, 113-134 (1992) · Zbl 0825.76422
[9]	Behr, M.; Johnson, A.; Kennedy, J.; Mittal, S.; Tezduyar, T., Computation of incompressible flows with implicit finite element implementations of the Connection Machine, Comput. Methods Appl. Mech. Engrg., 108, 99-118 (1993) · Zbl 0784.76046
[10]	Johan, Z.; Mathur, K. K.; Johnsson, S. L.; Hughes, T. J.R., An efficient communication strategy for finite element methods on the Connection Machine CM-5 system, Comput. Methods Appl. Mech. Engrg., 113, 363-387 (1994) · Zbl 0846.76046
[11]	Pothen, A.; Simon, H. D.; Liou, K.-P., Partitioning sparse matrices with eigenvectors of graphs, SIAM J. Matrix Analysis Appl., 11, 430-452 (1990) · Zbl 0711.65034
[12]	Simon, H. D., Partitioning of unstructured problems for parallel processing, Comput. Syst. Engrg., 2, 135-148 (1991)
[13]	Koelbel, C. H.; Loveman, D. B.; Schreiber, R. S.; Steele, G. L.; Zosel, M. E., (The High Performance Fortran Handbook (1994), MIT Press: MIT Press Cambridge, MA), ISBN 0-262-61094-9
[14]	Fortran, C. M., (Reference Manual (1994), Thinking Machines Corporation: Thinking Machines Corporation Cambridge, MA)
[15]	Tezduyar, T. E.; Behr, M.; Liou, J., A new strategy for finite element computations involving moving boundaries and interfaces — the deforming-spatial-domain/space-time procedure: I. The concept and the preliminary tests, Comput. Methods Appl. Mech. Engrg., 94, 339-351 (1992) · Zbl 0745.76044
[16]	Tezduyar, T. E.; Behr, M.; Mittal, S.; Liou, J., A new strategy for finite element computations involving moving boundaries and interface — the deforming-spatial-domain/space-time procedure: II. Computation of free-surface flows, two-liquid flows, and flows with drifting cylinders, Comput. Methods Appl. Mech. Engrg., 94, 353-371 (1992) · Zbl 0745.76045
[17]	Johnsson, S. L.; Mathur, K. K., Experience with the conjugate gradient method for stress analysis on a data parallel computer, Internat. J. Numer. Methods Engrg., 27, 523-546 (1989)
[18]	Johan, Z., Data parallel finite element techniques for large-scale computational fluid dynamics, (Ph.D. Thesis (1992), Department of Mechanical Engineering, Stanford University)
[19]	Mathur, K. K.; Johnsson, S. L., Communication primitives for unstructured finite element simulations on data parallel architectures, Comput. Syst. Engrg., 3, 63-71 (1992)
[20]	Shephard, M. S.; Georges, M. K., Automatic three-dimensional mesh generation by the finite octree technique, Internat. J. Numer. Methods Engrg., 32, 709-749 (1991) · Zbl 0755.65116
[21]	Tezduyar, T. E.; Mittal, S.; Ray, S. E.; Shih, R., Incompressible flow computations with stabilized bilinear and linear equal-order-interpolation velocity-pressure elements, Comput. Meth. Appl. Mech. Engrg., 95, 221-242 (1992) · Zbl 0756.76048
[22]	Kato, C.; Ikegawa, M., Large eddy simulation of unsteady turbulent wake of a circular cylinder using the finite element method, (Celik, I.; Kobayashi, T.; Ghia, K. N.; Kurokawa, J., Advances in Numerical Simulation of Turbulent Flows, FED, Vol. 117 (1991), ASME: ASME New York), 49-56