BIT Numerical Mathematics

, Volume 40, Issue 1, pp 158–176

Efficient Sparse LU Factorization with Left-Right Looking Strategy on Shared Memory Multiprocessors

  • O. Schenk
  • K. Gärtner
  • W. Fichtner
Article

DOI: 10.1023/A:1022326604210

Cite this article as:
Schenk, O., Gärtner, K. & Fichtner, W. BIT Numerical Mathematics (2000) 40: 158. doi:10.1023/A:1022326604210

Abstract

An efficient sparse LU factorization algorithm on popular shared memory multi-processors is presented. Pipelining parallelism is essential to achieve higher parallel efficiency and it is exploited with a left-right looking algorithm. No global barrier is used and a completely asynchronous scheduling scheme is one central point of the implementation. The algorithm has been successfully tested on SUN Enterprise, DEC AlphaServer, SGI Origin 2000 and Cray T90 and J90 parallel computers, delivering up to 2.3 GFlop/s on an eight processor DEC AlphaServer for medium-size semiconductor device simulations and structural engineering problems.

Parallel sparse LU factorization asynchronous computation scheduling SMP parallel computing multigrid coarse solver process simulation device simulation 

Copyright information

© Swets & Zeitlinger 2000

Authors and Affiliations

  • O. Schenk
    • 1
  • K. Gärtner
    • 2
  • W. Fichtner
    • 3
  1. 1.Integrated Systems Laboratory, Swiss Federal Institute of Technology ZürichETH ZürichZürichSwitzerland.
  2. 2.Weierstrass Institute for Applied Analysis and StochasticsBerlinGermany.
  3. 3.Integrated Systems Laboratory, Swiss Federal Institute of Technology ZürichETH ZürichZürichSwitzerland.

Personalised recommendations