Three-dimensional elasto-static analysis of 100 million degrees of freedom

Abstract GeoFEM, a parallel finite element analysis system, is application software that is being developed for conducting simulations of solid earth. This system is currently run on a massively parallel computer and can perform linear elastic analysis of a simple shape involving up to 100,000,000 degrees of freedom. In the present study, the memory capacity required and elapsed time were estimated for large-scale structural analyses, and the efficiencies were examined to investigate the parallel performance. This paper shows the possibility of performing super large-scale analyses with GeoFEM and a parallel computer.

[1]  Kengo Nakajima,et al.  Parallel Iterative Solvers with Localized ILU Preconditioning for Unstructured Grids on Workstation Clusters , 1999 .

[2]  Manolis Papadrakakis,et al.  Parallel solution techniques in computational structural mechanics , 1997 .

[3]  Genki Yagawa,et al.  A Large scale finite element analysis using domain decomposition method on a parallel computer , 1991 .

[4]  Charbel Farhat,et al.  A new finite element concurrent computer program architecture , 1987 .

[5]  Genki Yagawa,et al.  Study on Parallelization Method of Structural-Analysis Code , 1997, HPCN Europe.

[6]  Genki Yagawa,et al.  Porting an industrial sheet metal forming code to a distributed memory parallel computer , 1998 .

[7]  Genki Yagawa,et al.  A parallel finite element method with a supercomputer network , 1993 .

[8]  Charbel Farhat,et al.  A simple and efficient automatic fem domain decomposer , 1988 .

[9]  Genki Yagawa,et al.  GeoFEM: High-Performance Parallel FEM for Geophysical Applications , 1999, ISHPC.

[10]  O. Axelsson,et al.  On a class of preconditioned iterative methods on parallel computers. , 1989 .

[11]  Richard Barrett,et al.  Templates for the Solution of Linear Systems: Building Blocks for Iterative Methods , 1994, Other Titles in Applied Mathematics.

[12]  C. Farhat,et al.  The two-level FETI method. Part II: Extension to shell problems, parallel implementation and performance results , 1998 .

[13]  John N. Shadid,et al.  Parallel sparse matrix vector multiply software for matrices with data locality , 1998 .

[14]  Jacques Periaux,et al.  Domain decomposition methods for nonlinear problems in fluid dynamics , 1983 .

[15]  Jack J. Dongarra,et al.  Solving linear systems on vector and shared memory computers , 1990 .

[16]  C. Farhat,et al.  A method of finite element tearing and interconnecting and its parallel solution algorithm , 1991 .

[17]  Genki Yagawa,et al.  GeoFEM: High Performance Parallel FEM for Solid Earth , 1999, HPCN Europe.

[18]  Thomas J. R. Hughes,et al.  Scalability of finite element applications on distributed-memory parallel computers , 1994 .