At Tue, 19 Oct 2010 16:26:04 +0200, Thorsten Ottosen wrote:
Den 19-10-2010 14:59, Mathias Gaunard skrev:
On 18/10/10 23:41, David Abrahams wrote:
Just wanted to call some attention to this article I wrote about Boost.MPI:
http://daveabrahams.com/2010/09/03/whats-so-cool-about-boost-mpi/
I think it might be useful to link that from the Boost.MPI docs, too.
While the library seems nice (it bridges serialization with MPI), I don't really see anything in there that makes me think it's "so cool".
So after reading it, I felt a bit like the question "What's so cool about Boost.MPI?" remained unanswered.
The semi-automatic, but very easy to maintain, way of creating/maintaining the MPI type maps (via Boost.Serialization), IIUC.
Which yields huge performance benefits and makes practical some computations that might otherwise not be, due to resource constraints and the difficulty of generating type maps for arbitrary data structures. Do I need to be more explicit about that, or is it simply “not that cool?” -- Dave Abrahams BoostPro Computing http://www.boostpro.com