Basic HTML version of Foils prepared October 22 1997

Foil 101 Different Granularities of Decomposition I

From GCF Talk U - Lewis/NASA 3/9/92 Mardi Gras Conference on Concurrent Computing in Physical Sciences -- February 18, 1993. by Geoffrey C. Fox
We can divide problems and machines into interconnected modules - and we do this at different granularities We can consider two limiting cases (I) High performance Low Latency links Machine o Classic MPP such as Intel Paragon Problem o QCD: Components are gluon fields. Links are local. Software model o High Performance Fortran o Currently only regular structure o Eventually all such tightly coupled loosely synchronous problems o Fortran + Message passing

© 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 Wed Oct 22 1997