Minnesota Supercomputing Institute
2018.release, 2017/update3, 2018/release, 2019/update1
Thursday, January 10, 2019
The ScaLAPACK (or Scalable LAPACK) library includes a subset of LAPACK routines redesigned for distributed memory MIMD parallel computers. It is currently written in a Single-Program-Multiple-Data style using explicit message passing for interprocessor communication. It assumes matrices are laid out in a two-dimensional block cyclic decomposition. It provides routines for solving systems of simultaneous linear equations, least-squares solutions of linear systems of equations, eigenvalue problems, and singular value problems. Dense and banded matrices are handled, but not general sparse matrices.
On MSI systems ScaLAPACK libraries are now available primarily through the Intel MKL mathematical libraries. For more information please see the MKL webpage.
To use this software in a Linux environment run the command(s):
module load intel
Paths to the MKL libraries only (not including the Intel compilers) can be loaded with:
module load mkl
For information on how to compile using MKL libraries such as ScaLAPACK please see the MKL webpage.