Eigenvalue Problem Solvers - EPS

The Eigenvalue Problem Solver (EPS) is the main object provided by SLEPc. It is used to specify an eigenvalue problem, either in standard or generalized form, and provides uniform and efficient access to all of the eigensolvers included in the package.

Conceptually, the level of abstraction occupied by EPS is similar to other solvers in PETSc such as SLES for solving linear systems of equations.

EPS users can set various options at runtime via the options database (e.g., -eps_nev 4 -eps_type arnoldi ). Options can also be set directly in application codes by calling the corresponding routines (e.g., EPSSetDimensions()/ EPSSetType() ).

ex1.c: Solves a standard symmetric eigenproblem corresponding to the Laplacian operator in 1 dimension
ex2.c: Solves a standard symmetric eigenproblem corresponding to the Laplacian operator in 2 dimensions
ex3.c: Solves the same eigenproblem as in example ex2, but using a shell matrix
ex4.c: Solves an standard eigensystem Ax=kx with the matrix loaded from a file
ex5.c: Eigenvalue problem associated with a Markov model of a random walk on a triangular grid
ex7.c: Solves a generalized eigensystem Ax=kBx with matrices loaded from a file
ex8.c: Estimates the 2-norm condition number of a matrix A, that is, the ratio of the largest to the smallest singular values of A
ex9.c: Solves a problem associated to the Brusselator wave model in chemical reactions, illustrating the use of shell matrices
ex10.c: Illustrates the use of shell spectral transformations
ex11.c: Computes the smallest nonzero eigenvalue of the Laplacian of a graph
makefile