Intel® oneAPI HPC Toolkit
Get help with building, analyzing, optimizing, and scaling high-performance computing (HPC) applications.

Process tracing

youn__kihang
Novice
357 Views

Hello,

I am going to test the difference between mpirun and mpiexec on LSF.
The following manual says that mpirun is related to LSF.
(https://software.intel.com/en-us/mpi-developer-reference-linux-mpirun)

1. I would like to see what physical process mpirun and mpiexec use when working through LSF.
   Can the "-trace" command confirm which physical processes are being used?

2. Since there is a case where mpiexec is faster, I want to use mpiexec instead of mpirun.
    Is there a disadvantage of not being recognized by job scheduler?


Thanks

0 Kudos
4 Replies
PrasanthD_intel
Moderator
357 Views

Hi,

Thanks for reaching out to us.

 We will get back to you soon.

 

Prasanth

Anatoliy_R_Intel
Employee
357 Views

HI,

 

What do you mean by physical processes? Anyway, mpirun is just a wrapper for mpiexec to automatically detect job schedulers. Since Intel MPI 2019, mpiexec can also automatically detect job schedulers. In Intel MPI 2018 only mpirun could do this. So, now you can use just mpiexec.

--

Best regards, Anatoliy

Klaus-Dieter_O_Intel
357 Views

Please use "export I_MPI_DEBUG=5" in your LSF script to see which processors are used per node.

The flag "-trace" is used to create trace files for the Intel Trace Analyzer and Collector, see https://software.intel.com/en-us/trace-analyzer, i.e. to profile the MPI communication in your code.

Klaus-Dieter_O_Intel
357 Views

This issue is now closed. If you have a similar question, please start a new thread.

Reply