A mosaic preconditioner for a dual Schur complement

  • E. Tyrtyshnikov
  • Y. Vassilevski
Conference paper


The paper is devoted to designing an interface preconditioner for a Schur complement associated with Lagrange multipliers. We exploit the fact that underc ertain topological restrictions the inverted Schur complement is spectrally equivalent to a dense matrix whose entries are known explicitly. The preconditioner is a low-rank mosaic approximation of the dense matrix which is easy to evaluate on a vector.


Lagrange Multiplier Domain Decomposition Method Trace Normalization Topological Restriction Fast Multipole Method 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1] Scholar
  2. [2]
    Beatson, R., Greengard, L. (1997): A short course on fast multipole methods. In: Ainsworth, M. et al. (eds.): Wavelets, multilevel methods and elliptic PDEs. Oxford University Press, Oxford, pp. 1–37Google Scholar
  3. [3]
    Bramble, J., Pasciak, J., Schatz, A. (1989):The construction of preconditioners for elliptic problems by substructuring. IV. Math. Comp. 53, 1–24MathSciNetMATHGoogle Scholar
  4. [4]
    Bänsch, E. (1991): Local mesh refinement in 2 and 3 dimensions. Impact Cornput. Sci. Engrg. 3, 181–191MATHCrossRefGoogle Scholar
  5. [5]
    Farhat, C., Roux, F.-X. (1991): A method of finite element tearing and interconnecting and its parallel solution algorithm. Internal. J. Numer. Methods Engng. 32, 1205–1227MATHCrossRefGoogle Scholar
  6. [6]
    Farhat, C., Roux, F.-X. (1994): Implicit parallel processing in structural mechanics. Comput. Mech. Adv. 2, 1–124MathSciNetMATHCrossRefGoogle Scholar
  7. [7]
    Goreinov, S. (1999): Mosaic-skeleton approximations of matrices, generated by asymptotically smooth and oscillatory kernels. (Russian). In: Tyrtyshnikov, E. (ed.): Matrix methods and computations. Institute of Numerical Mathematics, Russian Academy of Sciences, Moscow, pp. 42–76Google Scholar
  8. [8]
    Hoppe, R., Iliash, Yu., Kuznetsov, Yu., Vassilevski, Yu., Wohlmuth, B. (1998): Analysis and parallel implementation of adaptive mortar element methods. East-West J. Numer. Math. 6, 223–248MathSciNetMATHGoogle Scholar
  9. [9]
    Hackbusch, W., Nowak, Z.P. (1989): On the fast matrix multiplication in the boundary element method by panel clustering. Numer. Math. 54, 463–491MathSciNetMATHCrossRefGoogle Scholar
  10. [10]
    Kiss, B., Krebsz, A. (1995): A sparse representation of the H 1/2 norm in finite element spaces. Preprint N 1. Department of Mathematics, Széchenyi Istvan College, Gyor, HungaryGoogle Scholar
  11. [11]
    Kuznetsov, Yu. (1995): Efficient iterative solvers for elliptic finite element problems on nonmatching grids. Russian J. Numer. Anal. Math. Modelling 10, 187–211MathSciNetMATHGoogle Scholar
  12. [12]
    Kuznetsov, Yu., Vassilevski, Yu. (2000): An interface preconditioner for the mortar element method. In: Neittaanmäki, P. et al. (eds.): Numerical mathematics and advanced applications. ENUMATH 99. World Scientific, Singapore, pp. 753–761Google Scholar
  13. [13]
    Rivara, M.-C. (1984): Mesh refinement processes based on the generalized bisection of simplexes. SIAM J. Numer. Anal. 21, 604–613MathSciNetMATHCrossRefGoogle Scholar
  14. [14]
    Nepomnyaschikh, S. (1991): Mesh theorems on traces, normalizations of function traces and their inversion. Soviet J. Numer. Anal. Math. Modelling 6, 223–242MathSciNetMATHGoogle Scholar
  15. [15]
    Nepomnyaschikh, S. (2000): Finite element trace theorems for parameter dependent Sobolev spaces. In: Neittaanmäki, P. et al. (eds.): Numerical mathematics and advanced applications. ENUMATH 99. World Scientific, Singapore, pp. 31–41Google Scholar
  16. [16]
    Tyrtyshnikov, E. (1997): Mosaic ranks and skeletons. In: Vulkov, L. et al. (eds.): Numerical analysis and its applications. (Lecture Notes in Computer Sciences Vol. 1196). Springer, Berlin, pp. 505–516CrossRefGoogle Scholar
  17. [17]
    Tyrtyshnikov, E. (1996): Mosaic-skeleton approximations. Calcolo 33, 47–57MathSciNetMATHCrossRefGoogle Scholar
  18. [18]
    Tyrtyshnikov, E. (2000): Incomplete cross approximation in the mosaic-skeleton method. Computing 64, 367–380MathSciNetMATHCrossRefGoogle Scholar
  19. [19]
    Vassilevski, Yu. (2001): A parallel interface preconditioner for the mortar element method in case of jumping coefficients. In: Chan, T. et al. (eds.): Domain decomposition methods in sciences and engineering., Augsburg, pp. 231–239Google Scholar

Copyright information

© Springer-Verlag Italia 2003

Authors and Affiliations

  • E. Tyrtyshnikov
    • 1
  • Y. Vassilevski
    • 1
  1. 1.Institute of Numerical MathematicsMoscowRussia

Personalised recommendations