A high resolution parallel Legendre transform algorithm

  • D. F. Snelling
Session 9B: Algorithms, Architectures And Performance II
Part of the Lecture Notes in Computer Science book series (LNCS, volume 297)


The spectral weather forecasting model at ECMWF is a sophisticated parallel program which uses global memory for the large shared data structure which contains the spectral representation of the atmosphere. In this paper the structure of this model is described and the section of the program which updates and references the shared spectral data structure is isolated. This extract, the Legendre transforms, contains all the critical parallel structure in the model. These transform algorithms are described and their relative performance on the Cray X-MP and the ETA 10 is presented.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    Louis J-F. ed., Research Manual 2, ECMWF Forecast Model, Adiabatic Part: a meteorological Bulletin (ECMWF, Reading: 1985).Google Scholar
  2. [2]
    Louis J-F. ed., Research Manual 3, ECMWF Forecast Model, Physical Parameterization: a Meteorological Bulletin (ECMWF, Reading: 1984).Google Scholar
  3. [3]
    Dent, D., The Multitasking Spectral Model at ECMWF, from Proceedings of Workshop on Using Multiprocessors in Meteorological Models (ECMWF, Reading: 1984).Google Scholar
  4. [4]
    Belousov, S.L., Tables of Normalized Associated Legendre Polynomials, (Pergamon Press, New York: 1962).Google Scholar
  5. [5]
    Daley, R. and Bourassa, Y., Rhomboidal versus Triangular Spherical Harmonic Truncation: some Verification Statistics., Atmosphere No. 16, 1978, p. 187–196.Google Scholar
  6. [6]
    Eliasen, E., Machenhauer, B., and Rasmussen, E., On a Numerical Method for Integration of the Hydrodynamic Equations with a Spectral Representation of Horizontal Fields. Rep. No. 2 Institut for Teoretisk Meteorologi, University of Copenhagen, 1970.Google Scholar
  7. [7]
    Machenhauer, B., and Rasmussen, E., On the Integration of the Spectral Hydrodynamic Equations by a Transform Method. Rep. No. 2 Institut for Teoretisk Meteorologi, University of Copenhagen, 1972.Google Scholar
  8. [8]
    Arnold, C.N., "Multitasking Library: Design Summary and Specification", ETA Internal document, July 1985.Google Scholar
  9. [9]
    Cray Research Inc., Programming Library Reference Manual, SR-0113, (Cray Research Inc., Mendota Heights, Minnesota), 1986, p. 14–1 to 14–31.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1988

Authors and Affiliations

  • D. F. Snelling
    • 1
  1. 1.European Centre for Medium-Range Weather ForecastsReadingEngland

Personalised recommendations