Difference between revisions of "MPI"

From HCL
Jump to: navigation, search
Line 11: Line 11:
 
MPI_Comm_free(&newcomm);
 
MPI_Comm_free(&newcomm);
 
}
 
}
<source>
+
</source>
 
Mind the overhead of <code>MPI_Comm_dup</code> and <code>MPI_Comm_free</code>.
 
Mind the overhead of <code>MPI_Comm_dup</code> and <code>MPI_Comm_free</code>.
  

Revision as of 16:44, 2 March 2010

Implementations

Tips & Tricks

  • For safe consecutive communications create new context, for example:
int communication_operation(MPI_Comm comm) {
MPI_Comm newcomm;
MPI_Comm_dup(comm, &newcomm);
... // work with newcomm
MPI_Comm_free(&newcomm);
}

Mind the overhead of MPI_Comm_dup and MPI_Comm_free.

Debugging

  • Add the following code:
int rank;
MPI_Comm_rank(MPI_COMM_WORLD, &rank);
if (!rank)
  getc(stdin);
MPI_Barrier(MPI_COMM_WORLD);
  • Compile your code with -g option
  • Run parallel application
  • Attach to process(es) from GDB
    • MPICH-1 runs a background process for each application process: 0, 0b, 1, 1b, ..., therefore, attach to the first ones.