ParalleloGAM: A parallel code for ODEs (Q1294511)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 1311268
Language Label Description Also known as
default for all languages
No label defined
    English
    ParalleloGAM: A parallel code for ODEs
    scientific article; zbMATH DE number 1311268

      Statements

      ParalleloGAM: A parallel code for ODEs (English)
      0 references
      0 references
      0 references
      19 July 1999
      0 references
      The method uses the ``parallelism across the steps''. It extends a previous paper of the authors [J. Comput. Appl. Math. 78, No. 2, 197-211 (1997; Zbl 0868.65039)] to nonlinear problems and is closely related to a companion paper of \textit{L. Brugnano} and \textit{D. Trigiante} [Appl. Numer. Math. 28, No. 1-4, 127-141 (1998; reviewed below)] for the theory. In a first coarse step the grid is determined (only 3 processors active). Then the linear system resulting from the Newton iteration is factored (block diagonal) and solved on \(p\) processors. Analysis of operations and numerical tests compared to a sequential Radau code. The parallel version wins only for higher accuracy (many steps).
      0 references
      numerical examples
      0 references
      parallel computation
      0 references
      parallelism across the steps
      0 references
      block boundary value methods
      0 references
      nonlinear
      0 references
      Newton iteration
      0 references

      Identifiers