Welcome to M.E.S.S.
MESS, the Matrix Equations Sparse Solvers library, is the successor to the Lyapack Toolbox for MATLAB®. It will be available as a MATLAB® toolbox as well as a C-library. It is intended for solving large sparse matrix equations as well as problems from model order reduction and optimal control. The C-version provides a large set of auxillary subroutines for sparse matrix computations and efficient usage of modern multicore workstations.
At the moment the following features are available or under active development:
- solving large scale (generalized) Lyapunov and Riccati equations.
- model order reduction for 1st and 2nd order systems via Balanced Truncation.
- H_2 model order reduction via the IRKA and TSIA algorithms.
- tools for LQR problems.
- Differential Riccati equation solvers.
The C-library additionally provides:
- a uniform interface to handle sparse and dense matrices.
- shared memory parallel algorithms using OpenMP and Pthreads.
- a solver for special sparse-dense Sylvester equations.
- a large set of direct and iterative solvers for linear systems.
- various helper subroutines for thread pools, configuration files, interface to MATLAB, ...
For information about the MATLAB-part please contact Dr. Jens Saak.
For information about the C-library please contact Martin Köhler.
We are collecting a number of common pitfalls regarding the compilation and usage on a separate troubleshooting page.
The C-library is tested nightly. The results are only available from our institute. See http://johannes/cdash/.