>So are you really talking about adding a whole new set of pt2pt and
>collectives APIs specifically to handle transactional communications? Or
>are you looking to extend the current APIs to have a mode that a user
>could set the error detection? I have to admit I still find the
>difference between the two modes that Rich describes not that far apart
>from each other.
I don't think that we're that far ahead on this but my personal
opinion is that we should not add new sets of MPI calls unless it is
truly necessary. I think that in this case we can get all the
functionality we need by adding
- a new function that the user would call at initialization time to
set the error detection/notification policy and
- communicating error events via new return values and error handler calls.
Greg Bronevetsky
Post-Doctoral Researcher
1028 Building 451
Lawrence Livermore National Lab
(925) 424-5756
bronevetsky1_at_[hidden]
|
![[Flag]](../../../../../images/flag-us.gif){kind=small}