Description: Previously, the use of matrix input would prohibit distributed memory parallel execution of the solver and element loop. This restriction has been removed. The matrix input portion of the analysis is handled by the root process, while the element loop and solver are distributed.