Basic HTML version of Foils prepared 27 October 1997

Foil 21 Parallelism in O(N2) N Body Approach I

From Master Set of Foils for GEM Computational Science Presentation GEM WorkShop Santa Fe -- 24-25 October 97. by Geoffrey C. Fox
1 O(N2) Algorithms are extremely efficient on parallel machines as
  • Load balancing guaranteed with equal numbers of particles in each node
  • Communication overhead goes like 1/(grain size n = number of particles in node) . Typical time to communicate word/ typical time to do floating computation) ~ 0.25(from algorithm details) . 10( = tcomm/tfloat)/n
  • So very efficient along as around 50 particles or more per processor
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 Mon Oct 27 1997