Title page for etd-0030118-154212


[Back to Results | New Search]

URN etd-0030118-154212
Author Chun-hung Chen
Author's Email Address as7218@gmail.com
Statistics This thesis had been viewed 5576 times. Download 0 times.
Department Applied Mathematics
Year 2017
Semester 1
Degree Master
Type of Document
Language English
Title Communication reduction problem in Schur complement method on distributed memory architecture.
Date of Defense 2017-06-29
Page Count 27
Keyword
  • large sparse linear system
  • distributed memory system
  • k-way partition
  • Schur complement
  • parallel computation
  • Abstract A common approach to solving a large sparse linear system in parallel is using the k-way
    partition method to relabel the variables and equations so that the rearranged matrix has a
    block arrow form, followed by Schur complement method to eliminate the resulting matrix
    to have a block upper triangular form. In the first step, the matrix is constituted by several
    independent submatrices so that they can be solved parallelly. However, if we do not
    carefully deal with the processes in calculation and storage, large mount of data communication
    between processes would occur in distributed memory system. In non-distributed
    memory system, the efficiency of parallelization might be low since the unbalanced loading
    of processes caused by the expensive calculation of forming Schur complement. In
    this paper, we will propose a novel method which can reduce the communication in the
    procedure of Schur complement simply by arranging the storagement and the method of
    calculation, so we might have better efficiency of the parallelization.
    Advisory Committee
  • Tzon-Tzer Lu - chair
  • Yueh-Cheng Kuo - co-chair
  • Chieh-Sen Huang - co-chair
  • Tsung-Lin Lee - advisor
  • Files
  • etd-0030118-154212.pdf
  • Indicate in-campus at 99 year and off-campus access at 99 year.
    Date of Submission 2018-01-31

    [Back to Results | New Search]


    Browse | Search All Available ETDs

    If you have more questions or technical problems, please contact eThesys