Preconditioned Krylov subspace methods solving dense nonsymmetric linear systems arising from BEM

Discretization of boundary integral equations leads, in general, to fully populated nonsymmetric linear systems of equations. We research the comparative performances of iterative techniques based on Krylov subspace solvers as GMRES(m), QMR and Bi-CGStab solving linear systems arising from BEM elasticity problems. Several general preconditioners are also considered and assessed. The results of numerical experiments suggest that preconditioned Krylov subspace methods are effective approaches for the solution of dense nonsymmetric linear systems arising from BEM.