Empirical study for communication cost of parallel conjugate gradient on a star-based network

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

Conjugate Gradient is an iterative linear solver that is used in many scientific and engineering applications to solve a system of linear equations. However, Conjugate Gradient generates a heavy load of computation and therefore it slows the performance of the applications using it. In this paper, we conduct an empirical cost study of a parallel CG on our star-based network. We evaluate the communication overhead involved by a parallel CG. In particular, we derive network parameters; the Maximum Transfer Unit (MTU), that can contribute to the optimization of communication cost and to the reduction of the waiting overhead of the parallel algorithm.

Original languageEnglish
Title of host publicationAMS2010
Subtitle of host publicationAsia Modelling Symposium 2010 - 4th International Conference on Mathematical Modelling and Computer Simulation
Pages498-503
Number of pages6
DOIs
Publication statusPublished - 2010
EventAsia Modelling Symposium 2010: 4th International Conference on Mathematical Modelling and Computer Simulation, AMS2010 - Kota Kinabalu, Borneo, Malaysia
Duration: May 26 2010May 28 2010

Publication series

NameAMS2010: Asia Modelling Symposium 2010 - 4th International Conference on Mathematical Modelling and Computer Simulation

Other

OtherAsia Modelling Symposium 2010: 4th International Conference on Mathematical Modelling and Computer Simulation, AMS2010
Country/TerritoryMalaysia
CityKota Kinabalu, Borneo
Period5/26/105/28/10

Keywords

  • Conjugate gradient
  • Cost study
  • Grid computing
  • Linear solvers
  • Parallel computing

ASJC Scopus subject areas

  • Applied Mathematics
  • Modelling and Simulation

Fingerprint

Dive into the research topics of 'Empirical study for communication cost of parallel conjugate gradient on a star-based network'. Together they form a unique fingerprint.

Cite this