Parallel Sparse Linear Algebra for Homotopy Methods

Simple item page
dc.contributor.authorDriver, Maria Sosonkina Jr.en
dc.contributor.committeechairWatson, Layne T.en
dc.contributor.committeememberJones, Mark T.en
dc.contributor.committeememberHeath, Lenwood S.en
dc.contributor.committeememberAllison, Donald C. S.en
dc.contributor.committeememberBeattie, Christopher A.en
dc.contributor.departmentComputer Scienceen
dc.date.accessioned2014-03-14T20:22:34Zen
dc.date.adate1997-09-19en
dc.date.available2014-03-14T20:22:34Zen
dc.date.issued1997-09-05en
dc.date.rdate1998-09-19en
dc.date.sdate1997-09-05en
dc.description.abstractGlobally convergent homotopy methods are used to solve difficult nonlinear systems of equations by tracking the zero curve of a homotopy map. Homotopy curve tracking involves solving a sequence of linear systems, which often vary greatly in difficulty. In this research, a popular iterative solution tool, GMRES(k), is adapted to deal with the sequence of such systems. The proposed adaptive strategy of GMRES(k) allows tuning of the restart parameter k based on the GMRES convergence rate for the given problem. Adaptive GMRES(k) is shown to be superior to several other iterative techniques on analog circuit simulation problems and on postbuckling structural analysis problems. Developing parallel techniques for robust but expensive sequential computations, such as globally convergent homotopy methods, is important. The design of these techniques encompasses the functionality of the iterative method (adaptive GMRES(k)) implemented sequentially and is based on the results of a parallel performance analysis of several implementations. An implementation of adaptive GMRES(k) with Householder reflections in its orthogonalization phase is developed. It is shown that the efficiency of linear system solution by the adaptive GMRES(k) algorithm depends on the change in problem difficulty when the problem is scaled. In contrast, a standard GMRES(k) implementation using Householder reflections maintains a constant efficiency with increase in problem size and number of processors, as concluded analytically and experimentally. The supporting numerical results are obtained on three distributed memory homogeneous parallel architectures: CRAY T3E, Intel Paragon, and IBM SP2.en
dc.description.degreePh. D.en
dc.identifier.otheretd-81897-131850en
dc.identifier.sourceurlhttp://scholar.lib.vt.edu/theses/available/etd-81897-131850/en
dc.identifier.urihttp://hdl.handle.net/10919/30718en
dc.publisherVirginia Techen
dc.relation.haspartmasha.pdfen
dc.rightsIn Copyrighten
dc.rights.urihttp://rightsstatements.org/vocab/InC/1.0/en
dc.subjectKrylov subspace methodsen
dc.subjectscientific computingen
dc.subjectiterative methodsen
dc.titleParallel Sparse Linear Algebra for Homotopy Methodsen
dc.typeDissertationen
thesis.degree.disciplineComputer Scienceen
thesis.degree.grantorVirginia Polytechnic Institute and State Universityen
thesis.degree.leveldoctoralen
thesis.degree.namePh. D.en

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
masha.pdf
Size:
931.94 KB
Format:
Adobe Portable Document Format
Download

Collections

Doctoral Dissertations