Advertisement

The Effect of Reordering Strategies on Rounding Errors in Large, Sparse Equation Systems

  • A. Ernst
  • W.-D. Schuh
Conference paper
Part of the International Association of Geodesy Symposia book series (IAG SYMPOSIA, volume 137)

Abstract

The effect of reordering strategies on the rounding errors is considered for the factorization and solution of sparse symmetric systems. On the one hand, a reduction of rounding errors can be expected, because the number of floating point operations decreases. On the other hand, the clustering of neighboring parameters and therefore the fixing of the sequence of parameter elimination may result in numerical instabilities. These effects are demonstrated for sparse covariance matrices in Wiener filtering. In particular Cholesky factorization and profile reordering in conjunction with envelope storage schemes are examined.

Keywords

Bouguer Anomaly Stochastic Approach Cholesky Factorization Binary Digit Linear Equation System 
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.

References

  1. Dahmen W, Reusken A (2008) Numerics for engineers and natural scientists (German). Springer, Berlin, HeidelbergGoogle Scholar
  2. Ernst A (2009) Implementation of efficient algorithms to reorder, solve and invert sparse normal equation systems with geodetic applications (German). Master’s thesis, Unitversity Bonn, Institute of Geodesy and GeoinformationGoogle Scholar
  3. Gaspari G, Cohn S, (1999) Construction of correlation functions in two and three dimensions. Q J Roy Meteorol Soc 125(554):723–757CrossRefGoogle Scholar
  4. Gaspari G, Cohn S, Guo J, Pawson S (2006) Construction and application of covariance functions with variable length-fields. Q J Roy Meteorol Soc, 132:1815–1838CrossRefGoogle Scholar
  5. Gibbs N, Poole W, Stockmeyer P (1976) An algorithm for reducing the bandwidth and profile of a sparse matrix. SIAM J Numer Anal 13:236–250CrossRefGoogle Scholar
  6. IEEE Computer Society (2008) IEEE standard for floating point arithmetic. Technical report, The Institute of Electrical and Electronics Engineers, IncGoogle Scholar
  7. Kreyszig E (1993) Advanced engineering mathematics, 7th edn. Wiley, New YorkGoogle Scholar
  8. Meissl P (1980) A priori prediction of roundoff error accumulation in the solution of a super-large geodetic normal equation system. NOAA/National Ocean Survey’s National Geodetic Survey (NGS), Rockville, Md., Professional Paper 12Google Scholar
  9. Moreaux G (2008) Compactly supported radial covariance functions. J Geodesy 82(7):431–443CrossRefGoogle Scholar
  10. Sansò F, Schuh W-D (1987) Finite covariance functions. Bulletin Géodésique 61:331–347CrossRefGoogle Scholar
  11. Schuh W-D (1991) Numerical behaviour of covariance matrices and their influence on iterative solution techniques. In: Rapp R, Sansò F (eds) Determination of the Geoid – Present and Future, IAG Proceedings, vol 106. Springer, Heidelberg, 432–441Google Scholar
  12. Snay R (1976) Reducing the profile of sparse symmetric matrices. NOAA Technical Memorandum, NOS NGS-4Google Scholar
  13. Stewart G (1973) Introduction to matrix computations. Academic press, New York, San Fancisco, LondonGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  1. 1.Institut für Geodäsie und Geoinformation der Universität BonnBonnGermany

Personalised recommendations