Hi Pavel, Lenny, Igor, list
Igor: Thanks for the pointer to your notes/paper!
Lenny: Thanks for resurrecting this thread!
Pavel: Thanks for the article!
It clarified a number of things about tuned collectives
(e.g. fixed vs. dynamic selection),
and the example rule file is very helpful too.
However, after reading the article and browsing through the code,
I still don't get what is that you call "segment size".
The article ranks "segment size" as an important parameter controlling
factor for good performance of collectives (section 5.1).
At first I thought "segment size" was the TCP/IP 'maximum segment size',
the MTU (or its IB equivalent) minus headers, etc.
However, the article apparently says it is not this (section 5.1).
What is "segment size" in OpenMPI?
Can the "segment size" be directly or indirectly controlled by the user?
On the other hand, the example rule file always has
topo and segmentation = 0.
Why?
Thank you,
Gus Correa
Pavel Shamis (Pasha) wrote:
Lenny,
You can find some details here:
http://icl.cs.utk.edu/news_pub/submissions/Flex-collective-euro-pvmmpi-2006.pdf
Pasha
Lenny Verkhovsky wrote:
Hi,
I am looking too for a file example of rules for dynamic collectives,
Have anybody tried it ? Where can I find a proper syntax for it ?
thanks.
Lenny.
On Thu, Jul 23, 2009 at 3:08 PM, Igor Kozin <i.n.ko...@googlemail.com
<mailto:i.n.ko...@googlemail.com>> wrote:
Hi Gus,
I played with collectives a few months ago. Details are here
http://www.cse.scitech.ac.uk/disco/publications/WorkingNotes.ConnectX.pdf
That was in the context of 1.2.6
You can get available tuning options by doing
ompi_info -all -mca coll_tuned_use_dynamic_rules 1 | grep alltoall
and similarly for other collectives.
Best,
Igor
2009/7/23 Gus Correa <g...@ldeo.columbia.edu
<mailto:g...@ldeo.columbia.edu>>:
> Dear OpenMPI experts
>
> I would like to experiment with the OpenMPI tuned collectives,
> hoping to improve the performance of some programs we run
> in production mode.
>
> However, I could not find any documentation on how to select the
> different collective algorithms and other parameters.
> In particular, I would love to read an explanation clarifying
> the syntax and meaning of the lines on "dyn_rules"
> file that is passed to
> "-mca coll_tuned_dynamic_rules_filename ./dyn_rules"
>
> Recently there was an interesting discussion on the list
> about this topic. It showed that choosing the right collective
> algorithm can make a big difference in overall performance:
>
> http://www.open-mpi.org/community/lists/users/2009/05/9355.php
> http://www.open-mpi.org/community/lists/users/2009/05/9399.php
> http://www.open-mpi.org/community/lists/users/2009/05/9401.php
> http://www.open-mpi.org/community/lists/users/2009/05/9419.php
>
> However, the thread was concentrated on "MPI_Alltoall".
> Nothing was said about other collective functions.
> Not much was said about the
> "tuned collective dynamic rules" file syntax,
> the meaning of its parameters, etc.
>
> Is there any source of information about that which I missed?
> Thank you for any pointers or clarifications.
>
> Gus Correa
>
---------------------------------------------------------------------
> Gustavo Correa
> Lamont-Doherty Earth Observatory - Columbia University
> Palisades, NY, 10964-8000 - USA
>
---------------------------------------------------------------------
> _______________________________________________
> users mailing list
> us...@open-mpi.org <mailto:us...@open-mpi.org>
> http://www.open-mpi.org/mailman/listinfo.cgi/users
>
_______________________________________________
users mailing list
us...@open-mpi.org <mailto:us...@open-mpi.org>
http://www.open-mpi.org/mailman/listinfo.cgi/users
------------------------------------------------------------------------
_______________________________________________
users mailing list
us...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users
_______________________________________________
users mailing list
us...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users
