Skip to Content.
Sympa Menu

charm - [charm] n_comm in LDStats

charm AT lists.cs.illinois.edu

Subject: Charm++ parallel programming system

List archive

[charm] n_comm in LDStats


Chronological Thread 
  • From: Roberto de Quadros Gomes <rqg.gomes AT gmail.com>
  • To: "charm AT cs.uiuc.edu" <charm AT cs.uiuc.edu>
  • Subject: [charm] n_comm in LDStats
  • Date: Fri, 16 Aug 2013 22:09:31 -0300
  • List-archive: <http://lists.cs.uiuc.edu/pipermail/charm/>
  • List-id: CHARM parallel programming system <charm.cs.uiuc.edu>

Hi,

I'm using a simple code written in AMPI for test and it have a loop like that:
"
i=100;
do
    {
        i--;
        res = MPI_Bcast(&sum, sizeof(int), MPI_INT, 0, MPI_COMM_WORLD);
        MPI_Barrier(MPI_COMM_WORLD);
        if( res != MPI_SUCCESS)
        MPI_Finalize();
       a--;
       if( a <= 0 )
       {
         MPI_Barrier(MPI_COMM_WORLD);
         MPI_Migrate();
         a = 5;
       }
   }while( i > 0);
"
When LB strategy is called, LDStats stats->n_comm always = 0. I believed this value would be 5 or larger than 5 because the MPI_Bcast function is there.
I had a problem like that when I tried read a pe_speed,  however I found out I needed use +LBTestPESpeed on command line. Is there any argument like this for LBStats to count each communication event?  Or am i doing confusing?



Archive powered by MHonArc 2.6.16.

Top of Page