Open MPI logo

Open MPI User's Mailing List Archives

  |   Home   |   Support   |   FAQ   |  

This web mail archive is frozen.

This page is part of a frozen web archive of this mailing list.

You can still navigate around this archive, but know that no new mails have been added to it since July of 2016.

Click here to be taken to the new web archives of this list; it includes all the mails that are in this frozen archive plus all new mails that have been sent to the list since it was migrated to the new archives.

Subject: Re: [OMPI users] OpenMPI collective algorithm selection
From: Ralph Castain (rhc_at_[hidden])
Date: 2013-04-05 09:11:51


You can get the headers installed by adding --with-devel-headers to the configure line.

On Apr 5, 2013, at 5:10 AM, chandan basu <cbasu_at_[hidden]> wrote:

> Hi,
>
> I want to use OpenMPI dynamic collective algorithm selection using rules file, e.g.
>
> mpirun --mca coll_tuned_use_dynamic_rules 1 --mca_coll_tuned_dynamic_rules_file rules.txt ./myexe
>
> I can see some examples in earlier discussions (given below). My question is how do I know the ID for the different collectives. I do not see coll_tuned.h in the installation folder . Is there any command to find the ID of an algorithm. I am particularly interested in Alltoallv. I have checked that coll_tuned_alltoallv_algorithm 1 and coll_tuned_alltoallv_algorithm 2 has lot of performance difference depending on data size and comm size. So I think giving a rules file can improve the performance over a range of data sizes and comm size. Any help in this regard will be appreciated.
>
> With regards,
>
> -Chandan
> ----------------------------------------
> Dr. Chandan Basu
> National Supercomputer Center
> Linköping University
> S-581 83 Linköping
> email: cbasu_at_[hidden]
> -----------------------------------------
> >1 # num of collectives
> >3 # ID = 3 Alltoall collective (ID in coll_tuned.h)
> >2 # number of com sizes
> >1 # comm size 1
> >1 # number of msg sizes 1
> >0 1 0 0 # for message size 0, linear 1, topo 0, 0 segmentation
> >8 # comm size 8
> >4 # number of msg sizes
> >0 1 0 0 # for message size 0, linear 1, topo 0, 0 segmentation
> >32768 2 0 0 # 32k, pairwise 2, no topo or segmentation
> >262144 1 0 0 # 256k, use linear 1, no topo or segmentation
> >524288 2 0 0 # message size 512k+, pairwise 2, topo 0, 0 segmentation
> ># end of first collective
>
>
> --
>
> _______________________________________________
> users mailing list
> users_at_[hidden]
> http://www.open-mpi.org/mailman/listinfo.cgi/users