On 2005-09-01 11:41 (-0500), Anthony Chan had pondered:
> On Thu, 1 Sep 2005, Chaitanya V. Hazarey wrote:
>
> > Hi all,
> >
> > I am a recent addition to this mailing list. We here working in the Grid
> > Computing Lab at TIFR, Mumbai. I am here trying to solve some popular
> > problems using MPI . This is done by using MPICH and LAM/MPI. Also tried
> > to get the LAM/MPI to work with Open PBS. But with moderate success.
> >
> > What I want to know is that how do we go about gathering profiling info
> > about the MPI Programs. And are there specfic tools avalable for the
> > same except the Vampir, whichare open source.
> >
> > Any pointers regarding MPI Code Instrumentation is most welcome.
>
> You can try MPE which comes with MPICHx. For LAM/MPI, you can download it
> from http://www.mcs.anl.gov/perfvis.
>
> Besides MPE, there is TAU which can be found at
> http://www.cs.uoregon.edu/research/tau/home.php
>
> A.Chan
Depending on what your exact objective is, you could also try the LAM
team's XMPI tool available off http://www.lam-mpi.org/software/xmpi
(guess it's not been updated for a while, but should still work)
Valgrind (http://valgrind.org/ ) is another tool that lets you also
profile code -- you might want to check if it'll let you profile MPI code
though.
Cheers, Arvind
_____________________________________________________________________
Arvind Gopu | High Performance Computing Group| (UITS-RAC-HPC) @ IU
HPC website: http://www.indiana.edu/~rac/hpc | Work: (812) 856-0187
My website: http://cs.indiana.edu/~agopu | Cell: (812) 361-4054
|
