Thanks for enlightening discussion. Sorry to be late.
However for the sake of clarity I have two more additional comments.
As I replied to Gus it was just a typo, I use to add to the mpirun
invokation
the full string --mca btl openib,self,sm.
However, as pointed out along the thread, I feel a little bit
unconfortable
with what the two mentioned FAQ seem to suggest to a newbie (like I feel
now!).
As far as concern the
http://www.open-mpi.org/faq/?category=tuning#selecting-components
only at the end is well stated that the esclusion of a componente (^tcp)
is equivalent to select all the others, at first sight, in the past, I
believed
that it ment use "openib" probably because I was thinking of my fabric
which gets only this two component. The other question which is clearly
stated (but a little bit hidden, IMHO) in the same sentence is:
"use only the self, sm, and openib components"
which finally says that all other component not included in the btl
parameter
are positively excluded.
Also the second mentioned FAQ:
http://www.open-mpi.org/faq/?category=openfabrics#ib-btl
is in my opinion misleading, as Gus pointed out.
While there is a special reference to the `self' component nothing is
said about the `sm' one.
Most of the users (and also the most part of my younger co-workers at
which this post is
specially dedicated) founding on the "Law of Least Astonishment" and
on the last lines fo the FAQ where is said:
Finally, note that if the openib component is available at run time,
Open MPI should automatically use it by default (ditto for self).
Hence, it's usuallyunnecessary to specify these options on the
mpirun command line. They are typically only used when you want to
be absolutely positively definitely sure to use the specific BTL.
assume that the `sm' flag is even included by default.
At the end, just for curiosity, as most part of applications show to
work with only
"openib,self" flags, what "physycally" happen to the "intranode" MPI
messages?
Thanks a lot
Salvatore Podda
Eugene Loh wrote:
> On 5/27/2011 4:32 AM, Jeff Squyres wrote:
>> On May 27, 2011, at 4:30 AM, Robert Horton wrote:
>>>> To be clear, if you explicitly list which BTLs to use, OMPI will
only
>>>> (try to) use exactly those and no others.
>>> It might be worth putting the sm btl in the FAQ:
>>>
>>> http://www.open-mpi.org/faq/?category=openfabrics#ib-btl
>> Is this entry not clear enough?
>>
>> http://www.open-mpi.org/faq/?category=tuning#selecting-components
> I think his point is that the example in the ib-btl entry would be
more
> helpful as a template for usage if it added sm. Why point users to a
> different FAQ entry (which we don't do anyhow) when three more
> characters ",sm" makes the ib-btl entry so much more helpful.
> _______________________________________________
> users mailing list
> users_at_[hidden]
> http://www.open-mpi.org/mailman/listinfo.cgi/users
Hi Jeff, list
I agree with Eugene and Robert.
By all means, please add ",sm" to "openib,self" in:
http://www.open-mpi.org/faq/?category=openfabrics#ib-btl
I am yet to see a situation where you want to run with openib and self,
but exclude sm (except for testing, perhaps when memcpy is broken).
Maybe that is what led Salvatore Podda think there was a
"Law of Least Astonishment" behind the mca parameters syntax,
which would insert "sm" automatically to the other two btl,
which is not really the case.
Like Salvatore, I've got confused by the mca parameter
syntax in the past also.
My recollection is that Jeff wrote the second
FAQ to placate my whining in the list about
to sm or not to sm.
However, the second FAQ clarifies the mca parameter logic,
along with the role of the "^" clause, and IMHO should be kept there:
http://www.open-mpi.org/faq/?category=tuning#selecting-components
My two cents,
Gus Correa
==================================================
Investi nel futuro. Investi nelle nostre ricerche.
Destina il 5 x 1000 all'ENEA
Cerchiamo:
- nuove fonti e nuovi modi per produrre energia pulita e sicura.
- modi migliori per utilizzare e risparmiare energia.
- metodologie e tecnologie per innovare e rendere piu' competitivo il sistema
produttivo nazionale.
- metodologie e tecnologie per la salvaguardia e il recupero dell'ambiente e
per la tutela della nostra salute e del patrimonio artistico del Paese.
Il nostro codice fiscale e': 01320740580