Open MPI logo

Hardware Locality Users' Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Hardware Locality Users mailing list

Subject: Re: [hwloc-users] MPI + threads parallelization
From: Samuel Thibault (samuel.thibault_at_[hidden])
Date: 2010-10-01 06:14:54

Ondrej Marsalek, le Fri 01 Oct 2010 10:56:42 +0200, a écrit :
> I have a question regarding parallelization using both MPI and threads
> in a NUMA setup. I would like to be able to use one MPI process per
> socket or NUMA node and use threads within each MPI process to use all
> the cores. Can I use hwloc to put and keep all of these in the right
> places?

Sure. You can for instance bind each whole MPI process to NUMA nodes and
let the system manage threads afterward, or even bind threads inside the
process. Of course, to get coherent things, you'll need to do a bit of
maths to bind according to the MPI rank number.

Note that lstopo --top shows the bound processes (and even threads on
Linux), which will probably useful to debug your code :)