Basic HTML version of Foils prepared February 13 00

Foil 20 General Analytical Form of Communication Overhead for Jacobi

From Master Set for Parallel Programming for Laplace's Equation CPS615 Spring Semester 00 -- February 00. by Geoffrey C. Fox
1 Consider N grid points in P processors with grain size n = N2/P
2 Sequential Time T1 = 4N2 tfloat
3 Parallel Time TP = 4 n tfloat + 4 ?n tcomm
4 Speed up S = P (1 - 2/N)2 / (1 + tcomm/(?n tfloat) )
5 Both overheads decrease like 1/?n as n increases
6 This ignores communication latency but is otherwise accurate
7 Speed up is reduced from P by both overheads
8 Load Imbalance Communication Overhead
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 Feb 21 2000