Basic HTML version of Foils prepared 15 March 1996

Foil 5 When will parallel computing take over?

From Complex Systems and Parallel Computing CPSP713 Case studies in Computational Science -- Spring Semester 1996. by Geoffrey C. Fox
When it has enough headroom (e.g. a factor of ten better "performance"/"cost-performance" than existing technology)
  • Definition of headroom confused as parallel machines much harder to use and so hard to decide on vlue to a given organization
nCUBE-1, CM-2 were comparable or perhaps a factor of 2 ... 4 better in cost-performance than conventional supercomputers
  • Enough to show "Parallel Computing Works"
  • not enough to "takeover"
DECmpp, IBM SP1, INTEL Paragon, CM-5, nCUBE-2 are (perhaps) a factor of ten better than competitive sequential (vector) machines
  • They can "takeover" ?

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 Sun Feb 22 1998