[Mpi-22] [MPI Forum] #96: MPI_OP_CREATE and associativity

[Torsten Hoefler]

On Tue, Jan 13, 2009 at 09:23:20AM +0100, Jesper Larsson Traeff wrote:
> maybe I misunderstand you, but I don't see that allowing operators to
> be explicitly non-associative would give you more freedom for optimizations.
> Doesn't your arguments above show that there are rather less than for 
> associative ops? But maybe it is better to discuss this at the next meeting?
just for the other readers: 
we clarified the discussion during the last Collectives Workinggroup
Teleconference. And decided to update ticket #96 and merge ticket #95
into it. 

For details see:
https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/96

> If your argument is: "there are natural (user-defined) operators in app 
> A, B and C that are mathematically not associative, and users want to perform 
> reductions with these" - then I agree that MPI is missing something.
yes, exactly. Such operators are MPI_SUM of MPI_FLOAT and friends.

> Maybe an "MPI_Op_create_nonassoc(...)"? (I also don't think an extra function
> for setting the associativity of an already defined function fits with MPI;
> although a function like MPI_Op_use_some_special_canonical_order(MPI_Op f),
> where f could be a user as well as a predefined MPI_Op,
> could perhaps make sense, and would allow to control the evaluation order 
> each time MPI_Reduce/MPI_Allreduce/... is called, but then such a function
> would either have to be collective or users would be required to use it
> consistently, ... that is, many complications)
exactly, this is why the whole discussion was moved from the context of
MPI-2.2 to MPI-3 into the collective working group. 

> Note also that non-associativity does put some non-trivial burden on 
> implementers: for non-associative (with or without hyphen?) operators, 
> a special algorithm that does the reduction in the sequential order is needed
I don't see a big/non-trivial problem.

I copied the collective working group list in order to take further
discussion off the MPI-2.2 list.

All the Best,
  Torsten


-- 
 bash$ :(){ :|:&};: --------------------- http://www.unixer.de/ -----
Torsten Hoefler       | Postdoctoral Researcher
Open Systems Lab      | Indiana University    
150 S. Woodlawn Ave.  | Bloomington, IN, 474045, USA
Lindley Hall Room 135 | +01 (812) 855-3608




* 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <http://lists.mpi-forum.org/pipermail/mpi-22/attachments/20090119/639b9b97/attachment.pgp>