Basic HTML version of Foils prepared 14 October 1997

Foil 45 Data Parallel Parallel Execution Time -II

From Fox Presentation Fall 1995 CPS615 Basic Simulation Track for Computational Science -- Fall Semester 95/96/97. by Nancy McCracken and Geoffrey C. Fox
1 Then the total time for the Runge-Kutta solver is:
2 Giving O(N2/Nproc) running time in the number of particles.
3 Note that parallel overhead or communication time/computation time is proportional to 1/n where n = N/Nproc is grain size.
4 Note this algorithm has an overhead characteristic of a one dimensional problem for
  • overhead goes like (1/n)1/d in d dimensions -- which is edge over area -- giving 1/n for d=1
in Table To:

© Northeast Parallel Architectures Center, Syracuse University, npac@npac.syr.edu

If you have any comments about this server, send e-mail to webmaster@npac.syr.edu.

Page produced by wwwfoil on Fri Oct 2 1998