Geodesic
On generalized relaxation method for linear saddle point problems
Matematičeskoe modelirovanie, Tome 13 (2001) no. 12, pp. 107-114 
E. V. Chizhonkov. On generalized relaxation method for linear saddle point problems. Matematičeskoe modelirovanie, Tome 13 (2001) no. 12, pp. 107-114. http://geodesic.mathdoc.fr/item/MM_2001_13_12_a11/
@article{MM_2001_13_12_a11,
     author = {E. V. Chizhonkov},
     title = {On generalized relaxation method for linear saddle point problems},
     journal = {Matemati\v{c}eskoe modelirovanie},
     pages = {107--114},
     year = {2001},
     volume = {13},
     number = {12},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/MM_2001_13_12_a11/}
}
TY  - JOUR
AU  - E. V. Chizhonkov
TI  - On generalized relaxation method for linear saddle point problems
JO  - Matematičeskoe modelirovanie
PY  - 2001
SP  - 107
EP  - 114
VL  - 13
IS  - 12
UR  - http://geodesic.mathdoc.fr/item/MM_2001_13_12_a11/
LA  - ru
ID  - MM_2001_13_12_a11
ER  - 
%0 Journal Article
%A E. V. Chizhonkov
%T On generalized relaxation method for linear saddle point problems
%J Matematičeskoe modelirovanie
%D 2001
%P 107-114
%V 13
%N 12
%U http://geodesic.mathdoc.fr/item/MM_2001_13_12_a11/
%G ru
%F MM_2001_13_12_a11

Voir la notice de l'article provenant de la source Math-Net.Ru

The preconditioned Arrow–Hurwicz algorithm is considered for solving a nonsingular system of the linear equations with a symmetric indefinite block matrix. Under the original assumption for invariant subspaces we solve an asymptotic optimization problem which depends on two spectral and two iterative parameters. For the optimal choice of iterative parameters the spectrum of the iteration operator is complex.