Open MPI logo

Open MPI Development Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Development mailing list

Subject: Re: [OMPI devel] RFC: meaning of "btl_XXX_eager_limit"
From: George Bosilca (bosilca_at_[hidden])
Date: 2009-07-23 16:07:54


On Jul 23, 2009, at 12:06 , Brian W. Barrett wrote:

> On Thu, 23 Jul 2009, Jeff Squyres wrote:
>
>> There are two solutions I can think of. Which should we do?
>>
>> a. Pass the (max?) PML header size down into the BTL during
>> initialization such that the the btl_XXX_eager_limit can
>> represent the max MPI data payload size (i.e., the BTL can size
>> its buffers to accommodate its desired max eager payload size,
>> its header size, and the PML header size). Thus, the
>> eager_limit can truly be the MPI data payload size -- and easy
>> to explain to users.
>
> This will not work. Remember, the PML IS NOT THE ONLY USER OF THE
> BTLS. I'm really getting sick of saying this, but it's true. There
> can be no PML knowledge in the BTL, even if it's something simple
> like a header size. And since PML headers change depending on the
> size and type of message, this seems like a really stupid parameter
> to publish to the user.

Without talking about the fact that each PML have it's own headers
(and corresponding sizes). Do we really expect the BTLs (or something
else) to compute the maximum as a convenience for the user? Without
forcing the user to select the desired PML, we will not be able to
provide an accurate length for the eager.

>
>> b. Stay with the current btl_XXX_eager_limit implementation (which
>> OMPI has had for a long, long time) and add the code to check
>> for btl_eager_limit less than the pml header size (per this past
>> Tuesday's discussion). This is the minimal distance change.
>
> Since there's already code in Terry's hands to do this, I vote for b.

Or prevent the users to play with something they are not really
supposed to play with. My point is that the eager is not related to
MPI, it is a BTL property based on the hardware, and should only be
tweaked by someone who understand what's he/she's playing with.

The only interest I can see in this RFC is to allow developer/users to
play with real ZERO bytes eager messages. While this is not
performance thing (at least I don't think that in a normal/correct MPI
application this can improve the performance) such an option can be a
life-saver for applications generating a lot of unexpected messages.

Otherwise, I totally agree with Brian's comments.

george.

>
>> 2. OMPI currently does not publish enough information for a user to
>> set eager_limit to be able to do BTL traffic shaping. That is, one
>> really needs to know the (max) BTL header length and the (max) PML
>> header length values to be able to calculate the correct
>> eager_limit force a specific (max) BTL wire fragment size. Our
>> proposed solution is to have ompi_info print out the (max) PML and
>> BTL header sizes. Regardless of whether 1a) or 1b) is chosen, with
>> these two pieces of information, a determined network administrator
>> could calculate the max wire fragment size used by OMPI, and
>> therefore be able to do at least some of traffic shaping.
>
> Actually, there's no need to know the PML header size to shape
> traffic. There's only need to know the BTL header, and I wouldn't be
> opposed to changing the behavior so that the BTL eager limit
> parameter included the btl header size (because the PML header is
> not a factor in determining size of individual eager packets). It
> seems idiotic, but whatever - you should more care about what the
> data size the user is sending than the MTU size. Sending multiple
> MTUs should have little performance on a network that doesn't suck
> and we shouldn't be doing all kinds of hacks to support networks
> who's designers can't figure out which way is up.
>
> Again, since there are multiple consumers of the BTLs, allowing
> network designers to screw around with defaults to try and get what
> they want (even when it isn't what they actually want) seems
> stupid. But as long as you don't do 1a, I won't object to
> uselessness contained in ompi_info.
>
> Brian
> _______________________________________________
> devel mailing list
> devel_at_[hidden]
> http://www.open-mpi.org/mailman/listinfo.cgi/devel