Xing Feng, A more focused (and certainly more detailed) analysis of the cost of different algorithms for collective communications can be found in [1], and more recently in [2].
George. [1] http://icl.cs.utk.edu/projectsfiles/rib/pubs/Pjesivac-Grbovic_PMEO-PDS05.pdf [2] https://www.cs.utexas.edu/~echan/vpapers/CCPE2007.pdf On Wed, Sep 30, 2015 at 3:07 AM, Marc-Andre Hermanns < m.a.herma...@grs-sim.de> wrote: > Dear Xing Feng, > > there are different algorithms to implement collective communication > patterns. Next to general Big-O complexity the concrete complexity > also depends on the network topology, message length, etc.. > > Therefore many MPI implementations switch between different algorithms > depending on the concrete communication parameters in a call. > > A colleague of mine investigated some MPI implementations (though not > OpenMPI) [1]. There you can see how different MPI implementations > (IBM, ParaStation, Cray) scale differently for a selection of > collective calls. Maybe that helps a little in understanding the > performance of your application. > > Cheers, > Marc-Andre > > [1] http://dl.acm.org/citation.cfm?doid=2751205.2751216 > > > On 30.09.15 07:53, XingFENG wrote: > > Hi, every one, > > > > I am working with open-mpi. When I tried to analyse performance of my > > programs, I find it is hard to understand the communication complexity > > of MPI routines. > > > > I have found some page on Internet such as > > http://stackoverflow.com/questions/10625643/mpi-communication-complexity > > > > > > This indicates that communication complexity of broadcasting an > > integer is O(log P) where P is the number of processes. But is it > > correct on different MPI implementations( OMPI, MPICH, etc.)? Is there > > an official document discussing such complexity? > > > > > > -- > > Best Regards. > > --- > > Xing FENG > > PhD Candidate > > Database Research Group > > > > School of Computer Science and Engineering > > University of New South Wales > > NSW 2052, Sydney > > > > Phone: (+61) 413 857 288 > > > > > > _______________________________________________ > > users mailing list > > us...@open-mpi.org > > Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users > > Link to this post: > http://www.open-mpi.org/community/lists/users/2015/09/27719.php > > > > -- > Marc-Andre Hermanns > Jülich Aachen Research Alliance, > High Performance Computing (JARA-HPC) > German Research School for Simulation Sciences GmbH > > Schinkelstrasse 2 > 52062 Aachen > Germany > > Phone: +49 2461 61 2509 > Fax: +49 241 80 6 99753 > www.grs-sim.de/parallel > email: m.a.herma...@grs-sim.de > > > _______________________________________________ > users mailing list > us...@open-mpi.org > Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/users > Link to this post: > http://www.open-mpi.org/community/lists/users/2015/09/27720.php >
