On algebraic multi-level methods for non-symmetric systems --- comparison results (Q952047)

Algebraic multi-level methods are applied to linear systems with non-symmetric \(M\)-matrices. The authors think of applications with Markov chain modeling and other problems in probability, economics, operations research, and the model problem of elliptic differential equations. A block factorization of the matrix is assumed in the sense of multilevel methods. The AMLI, the MAMLI, and the symmetric AMLI procedure are investigated and a theoretical comparison between the AMG methods and the classical block Jacobi and block Gauss-Seidel methods is established.