Browse > Article
http://dx.doi.org/10.12989/sem.2002.14.6.625

Automatic decomposition of unstructured meshes employing genetic algorithms for parallel FEM computations  

Rama Mohan Rao, A. (Structural Engineering Research Centre, CSIR Campus)
Appa Rao, T.V.S.R. (Structural Engineering Research Centre, CSIR Campus)
Dattaguru, B. (Department of Aerospace Engineering, Indian Institute of Science)
Publication Information
Structural Engineering and Mechanics / v.14, no.6, 2002 , pp. 625-647 More about this Journal
Abstract
Parallel execution of computational mechanics codes requires efficient mesh-partitioning techniques. These mesh-partitioning techniques divide the mesh into specified number of submeshes of approximately the same size and at the same time, minimise the interface nodes of the submeshes. This paper describes a new mesh partitioning technique, employing Genetic Algorithms. The proposed algorithm operates on the deduced graph (dual or nodal graph) of the given finite element mesh rather than directly on the mesh itself. The algorithm works by first constructing a coarse graph approximation using an automatic graph coarsening method. The coarse graph is partitioned and the results are interpolated onto the original graph to initialise an optimisation of the graph partition problem. In practice, hierarchy of (usually more than two) graphs are used to obtain the final graph partition. The proposed partitioning algorithm is applied to graphs derived from unstructured finite element meshes describing practical engineering problems and also several example graphs related to finite element meshes given in the literature. The test results indicate that the proposed GA based graph partitioning algorithm generates high quality partitions and are superior to spectral and multilevel graph partitioning algorithms.
Keywords
load balancing; mesh partitioning; genetic algorithms; multilevel approaches; unstructured meshes; parallel processing; dual graph;
Citations & Related Records

Times Cited By Web Of Science : 1  (Related Records In Web of Science)
Times Cited By SCOPUS : 5
연도 인용수 순위
1 Barnard, S.T. and Simon, H.D. (1994), "A fast multilevel implementation of recursive spectral bisection forpartitioning unstructured problems", Concurrency: Practice and Experience, 6, 101-107.   DOI   ScienceOn
2 Bouhmala, N. and Pahud, M. (1998), "A parallel variant of simulated annealing for optimizing mesh partitionson workstations", Advances in Engineering Software, 29(3-6), 481-485.   DOI   ScienceOn
3 Garey, M.R. and Johnson, D.S. (1979), Computers and Intractability: A guide to the theory of NP-Completeness,Freeman, W.H. and Company, N.Y.
4 Mitchel, W.J., Steadman, J.P. and Liggett, R.S. (1976), "Synthesis and optimisation of small rectangular floorplans", Environment and Planning, 3, 37-70.   DOI
5 Park, K. and Carter, B. (1995), "On the effectiveness of genetic search in combinatorial optimisation", Proc. the10th ACM Symposium on Applied Computing, Genetic Algorithms and Optimisation Track.
6 Soper, A.J., Walshaw, C. and Cross, M. (2000), "A combined evolutionary search and multilevel optimisationapproach to graph partitioning", Mathematics Research Report 00/IM/58.
7 Walshaw, C. and Cross, M. (1999), "Parallel optimisation algorithms for multilevel mesh partitioning",Mathematics Research Report 99/IM/44.
8 Williams, R.D. (1991), "Performance of dynamic load balancing algorithms for unstructured mesh calculations",Concurrency, 3, 457-481.   DOI
9 Rama Mohan Rao, A., Umesha, P.K. and Loganathan, K. (1998), "PSTRAIN: A graphic user interface forparallel nonlinear dynamic analysis of structures", SERC Technical Report NO. GAP 0641-PPG-TR-98-02.
10 Karypis, G. and Vipin Kumar. (1999), "A fast and high quality multilevel scheme for partitioning irregular graphs", SIAM Journal of Scientific Computing, 20(1), 359-392.   DOI   ScienceOn
11 Kernighan, B.W. and Lin, S. (1970), "An efficient heuristic procedure for partitioning graphs", The Bell SystemTechnical Journal, 29(2), 291-307.
12 De Keyser, J. and Roose, D. (1992), "Grid partitioning by inertial recursive bisection", Technical Report TW174, Department of Computer Science, Belgium.
13 Goldberg, D. and Segrest, P. (1987), "Finite Markov chain analysis of genetic algorithms", Proc. the Fifth Int.Conf. on Genetic Algorithms.
14 Mansour, N. and Fox, G.C. (1994), "Allocating data to distributed memory multiprocessors by geneticalgorithms", Concurrency: Practice and Experience, 6(6), 485-504.   DOI   ScienceOn
15 Punch, W., Goodman, E., Pei, M., Lai, C.S., Hovland, P. and Enbody, R. (1993), "Intelligent clustering of highdimensionality data using genetic algorithms", Proc. Int. Conf. on Genetic Algorithms, 557-564 .
16 Miller, G.L., Teng, S., Thurston, W. and Vavasis, S.A. (1998), "Geometric separators for finite element meshes",SIAM Journal of Scientific Computing, 19(2), 364-386.   DOI   ScienceOn
17 Khan, A.I. and Topping, B.H.V. (1998), "Subdomain generation for parallel finite element analysis", ComputingSystems in Engineering, 4(4-6), 96-129.
18 Wendl, S. (1996), "A seed based decomposition algorithm employing genetic algorithms", Report No. EPCCSS96-01, University of Edinburgh, UK.
19 Holland, J. (1975), Adaptation in Natural and Artificial Systems, University of Michigan Press. Ann Arbor. MI.
20 Altman, E.R., Agarwal, V.K. and Rao, G.R. (1993), "A novel methodology using genetic algorithms for the design of caches and cache replacement policy", Proc. Int. Conf. on Genetic Algorithms, 392-399.
21 Rama Mohan Rao, A., Appa Rao, T.V.S.R. and Dattaguru, B. (1998), "Load balancing for parallel finite elementanalysis employing artificial neural networks", In CDROM Proc. Int. Conf. on Theoretical AppliedComputational and Experimental Mechanics.
22 Hendrickson, B. and Leland, R. (1995), "A multilevel algorithm for partitioning graphs", Proc. theSupercomputing 95, ACM.
23 Gil, C., Ortega, J., Diaz, A.F. and Monotoya, M.G. (1998), "Annealing based heuristics and genetic algorithmsfor circuit partitioning in parallel test generation", Future Generation Computing Systems, 14(5), 439-451.   DOI   ScienceOn
24 Farhat, C. (1988), "A fast simple and efficient automatic FEM domain decomposer", Computers and Structures,28(5), 579-602.   DOI   ScienceOn
25 Beasley, D., Bull, D.R. and Martin, R.R. (1993), "An overview of genetic algorithms: part 2, research topics",University Computing, 15(4), 170-181.
26 Goldberg, D.E. (1989), Genetic Algorithms in Search, Optimisation and Machine Learning Addison, Wesley, N.J.
27 Jones, M.T. and Plassmann, P.E. (1994), "Computational results for parallel unstructured mesh computations",Computing Systems in Engineering, 5(4-6), 297-309.   DOI   ScienceOn
28 Mansoor, N. and Fox, G.C. (1991), "A hybrid genetic algorithm for task allocation in multi-computers", Proc.the 4th Int. Conf. on Genetic Algorithms (Ed. Belew, R.K. and Booker, L.B.), Morgan Kaufmann, 466-473.
29 Plaskacz, E.J., Ramirez, M.R. and Gupta, S. (1994), "Non-linear explicit transient finite element analysis on theintel delta", Computing Systems in Engineering, 5, 1-17.   DOI   ScienceOn
30 Rama Mohan Rao, A. (2001), "Efficient parallel processing algorithms for nonlinear dynamic analysis", Ph.D.Thesis Submitted to Indian Institute of Science, Bangalore. India.
31 Simon, H.D. (1991), "Partitioning of unstructured problems for parallel processing", Computing Systems inEngineering, 2, 135-148.   DOI   ScienceOn
32 Pain, C.C., De Oliveira, C.R.E. and Goddard, A.J.H. (1999), "A neural network graph partitioning procedure forgrid based domain decomposition", Int. J. Numer. Methods Eng., 44, 593-613.   DOI   ScienceOn