On Jul 11, 2007, at 9:57 PM, Pedro Petrovitch wrote:
> Altough my question isnt directly about LAM/MPI it kind of answers
> me a lot
> of things that I need to know if I need to use MPI or not. Here it
> goes: Is
> it possible to make a parallel algorithm (using MPI) running in a
> single
> machine/node (with many process running on it, i.e., mpirun -np 10
> main and
> only one hardware processor) to run faster than a serial one on the
> same
> conditions? Thanks a lot for the atention. Any help would be
> appreciated.
Not generally. If it did, chances are likely that you'd actually just
be abusing the scheduler into giving your user more CPU cycles than
you'd otherwise (i.e. with fewer processes) receive.
Normally though, even ignoring MPI-library overhead, you're simply
doing the same thing you'd do in serial except with the addition of
lots of unnecessary process swapping and extra memory copying
(between MPI processes) that you wouldn't need to do if everything
were in serial mode.
~Kyle
|
