Parallel Computing in Computational Mechanics

Zdeněk Bittnar, Jaroslav Kruis, Jiří Němeček, Bořek Patzák, Daniel Rypl

Department of Structural Mechanics
Faculty of Civil Engineering
Czech Technical University in Prague
Thákurova 7, 166 29 Prague, Czech Republic



Abstract:

The Finite Element Method is a powerful numerical method for solving partial differential equations. It is widely used in many fields in civil, mechanical, biomechanical, aerospace, and electrical engineering. Using this method the engineer can simulate the thermo-chemo-hydro mechanical behaviour of solids, fluids, and structures. In many cases, these calculations are frequently very time demanding, especially when the underlying models are non-linear and three-dimensional. Typical examples are the simulation of crack growth which requires up to several weeks of computing time, and most of the problems in multi-physics. The typical trouble in these cases is ill-conditioning of governing equations. Since several of these simulations are required to evaluate a numerical model or structural design, it is necessary to speed up the computations. An attractive way to achieve this speed up is the use of parallel algorithms.