A Parallel Approach for Solving a Wide Class of Structural Non-Linear Problems

The joint use of powerful algorithms and parallel computers is necessary to strongly reduce the numerical cost of complex simulations taking into account strongly non-linear material models and geometrical non-linearities. Those simulations are required in order to obtain accurate numerical predictions, especially to respect safety conception constrains which are more and more required in high-tech industries. We present the principles of a parallel approach suited to the simulation of a wide class of non-linear problems. It is based, on the one hand, on the use of two domain decompositions, in order to make use of the mechanical properties of the different type of equations to be solved; and on the other hand, on the design of a full-featured language, which allows the user to distribute tasks on a parallel machine. To facilitate the resolution in parallel of non-linear problems, in a transparent way for the user while ensuring a good effectiveness, various recent developments were carried out. The asynchronous execution of calculations at the user level was simplified, on the one hand, using a "container" object gathering the decompositions of the objects used by a calculation; and on the other hand, using an operator to distribute the tasks on various processors starting from these decompositions. Moreover the resolution of large scale problems requires an intensive use of the virtual memory (swap on disk for unused objects); thus, the management of objects which can be shared between various applications has been optimised in order to ensure the data coherence and to limit the blocking phases of the parallel applications. A version based on the standard Posix "pthread" was initially developed to ensure the performances of the parallel programming environment and to allow the code portability on shared memory computers. The extension of this strategy to shared-distributed memory computers is under way.