commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From S├ębastien Brisard <sebastien.bris...@m4x.org>
Subject Re: [Math] Moving on or not?
Date Fri, 08 Feb 2013 19:04:12 GMT
Hello,


2013/2/6 Konstantin Berlin <kberlin@gmail.com>

> >
> > As for efficiency (or faster execution, if you want), I don't see the
> > point in doubting that tasks like global search (e.g. in a genetic
> > algorithm) will complete in less time when run in parallel...
>
>
> Just a quick note. This statement is incorrect. Parallelization should be
> done at the coarsest level. So if the program is already parallel, and each
> thread is executing your parallel math function, it could indeed be worse.
>

I think *this* statement is incorrect. Whether or not you need
parallelization is pretty much dependent on the application. For example,
if I'm dealing with a stack of 2D images, having a single thread
implementation is probably best. That way, I can decide to treat different
images in different threads (parallelization would then occur at the
coarsest level). On the other hand, if I'm dealing with a 3D image, maybe I
would appreciate that each operation is parallelized. If that's not done at
the finest level, then the user is stuck with a computation which takes for
ever.

S

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message