That sounds about right. Some of the names for the BGQ-specific optimized collectives are pretty elaborate. SCATTER_MPICH Is tame by comparison :-) I can't remember if the names are listed in the redbook, but you can always grep the source code. Jeff On Thu, Nov 19, 2015 at 8:24 PM, Jen B <[email protected]> wrote:
Thank you for your answers and helpful suggestions, Rob and Jeff!
I have my last question I hope someone could help me clarify. So I get algorithm names for MPI collectives. Is there any convention for them? I did not find this information in the redbook. For example, for some of the algorithm names I get "SCATTER_MPICH", so I think this means that if the algorithm name ends with "_MPICH" then this is using the algorithm from MPICH, not the ones optimized for BG/Q. Is this correct?
Thank you! Jen
On Thu, Nov 19, 2015 at 5:49 PM, Jeff Hammond <[email protected]> wrote:
As Rob said, you will find lots of info in the redbook. You can set all sorts of debug options to print out the algorithm(s) used on every call, or prescribe them explicitly (not recommended).
Jeff
On Thu, Nov 19, 2015 at 8:38 AM, Jen B <[email protected]> wrote:
Thank you for your quick response, Jeff!
I was wondering the values for "short" and "large" message sizes, they've been used throughout the paper but not mentioned explicitly. Do you happen to know if these values are explicitly mentioned somewhere?
Also, if I get the last algorithm name used for the MPI collective routine, is there a way I can for sure tell which algorithm is being used? For example, is there any name formatting convention that is used.
Sorry for so many questions. Any help with any of these questions will be greatly appreciated!
Thank you! Jen
On Wed, Nov 18, 2015 at 6:06 PM, Jeff Hammond <[email protected]> wrote:
A binary tree is the implementation of last resort.
See these papers for details of the MPI collective implementations on Blue Gene: - http://dx.doi.org/10.1177/1094342014552086 - http://dx.doi.org/10.1109/HOTI.2009.12 (this is not an inclusive list)
http://lists.mpich.org/pipermail/discuss/2015-March/003754.html has some related info.
Jeff
On Wed, Nov 18, 2015 at 12:53 PM, Jen B <[email protected]> wrote:
Hello,
I have a couple of questions about BG/Q collectives and would appreciate any help. I was wondering if Do IBM BG/Q MPI collectives use binomial tree? Once I find the name of the algorithm used by the MPI collective routine using MPIX_Get_last_algorithm_name, is there a quick reference or guide to check how it is implemented?
Thanks! Jen
_______________________________________________ discuss mailing list [email protected] To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss
-- Jeff Hammond [email protected] http://jeffhammond.github.io/
_______________________________________________ discuss mailing list [email protected] To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss
_______________________________________________ discuss mailing list [email protected] To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss
-- Jeff Hammond [email protected] http://jeffhammond.github.io/
_______________________________________________ discuss mailing list [email protected] To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss
_______________________________________________ discuss mailing list [email protected] To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss
-- Jeff Hammond [email protected] http://jeffhammond.github.io/ _______________________________________________ discuss mailing list [email protected] To manage subscription options or unsubscribe: https://lists.mpich.org/mailman/listinfo/discuss