I am new to LAPACK and am starting a project that requires potentially large matrix decompositions and solves. What I'm looking to do right now is to solve a simple problem of Ax = B in C using ScaLAPACK. This will later be used to solve a much harder problem. I should be running in a Unix Cluster, but I won't be worrying about exploiting parallelism or anything at the moment.

From my understanding, the best approach is to use gaussian elimination and matmul/Dgemm. Is the gaussian elimination relatively fast?

Is the install just the standard way like in http://www.netlib.org/scalapack/. Basically, I am unsure if it will work seamlessly in C?

I'm just getting started, but I hope to start picking it up myself soon. Any help is appreciated. Thanks.