I apologize if this is a basic question, but I am looking for an implementation of parallel (distributed) matrix-matrix multiplication kernels, as well as LU decomposition. Does one of the Trilinos packages have an interface to these? If so, which one would it be? Thanks Raj