An MPI parallel implementation of Newmark's method

The standard message-passing interface (MPI) is used to parallelize Neu mar k's method The linens matrix equation encountered at each time step is solv ed using a preconditioned conjugate gradient algorithm. Data are distribute d over the processors of a given parallel computer on a degree-of-freedom b asis; this produces effective load balance between the processors and leads to a highly parallelized code. The portability of the implementation of th is scheme is tested by solving some simple problems on two different machin es: an SGI Origin2000 and an IBM SP2. The measured times demonstrate the ef ficiency of the approach and highlight the maintenance advantages that aris e from using a standard parallel library such as MPI.