Skip to Content

Using Scalapack

ScaLAPACK is a library containing a subset of LAPACK routines that have been redesigned to run on distributed-memory MIMD parallel computers.

ScaLAPACK is written in a Single-Program-Multiple-Data style that uses explicit message passing for interprocessor communication. It assumes that matrices are laid out in a two-dimensional block cyclic decomposition.

More information about using ScaLAPACK is available from its home page.

Code Examples

We will add back some code examples from the old cluster's documentation once we have had a chance to verify that they work on the Amber cluster.

In the meantime, you can try the official examples from the ScaLAPACK site.