Parallel -Matrix Arithmetics on Shared Memory Systems

Ronald Kriemann

Contact the author: Please use for correspondence this email.
Submission date: 12. May. 2004
Pages: 22
published in: Computing, 74 (2005) 3, p. 273-297 
DOI number (of the published article): 10.1007/s00607-004-0102-2
Bibtex
MSC-Numbers: 65F05, 65F30, 65Y05, 65Y10, 68W10
Keywords and phrases: hierarchical matrices, parallel algorithms, shared memory systems
Download full preprint: PDF (467 kB), PS ziped (371 kB)

Abstract:
-matrices, as they were introduced in previous papers, allow the usage of the common matrix arithmetic in an efficient, almost optimal way. This article is concerned with the parallelisation of this arithmetics, in particular matrix building, matrix-vector multiplication, matrix multiplication and matrix inversion.

Of special interest is the design of algorithms, which reuse as much as possible of the corresponding sequential methods, thereby keeping the effort to update an existing implementation at a minimum. This could be achieved by making use of the properties of shared memory systems as they are widely available in the form of workstations or compute servers. These systems provide a simple and commonly supported programming interface in the form of POSIX-Threads.

The theoretical results for the parallel algorithms are tested with numerical examples from BEM and FEM applications.