[Mpi3-ft] MPI RMA FT Notes

[Wesley Bland]

Jim and I have drafted some text based on these notes. Rather than push the entire PDF back to the mailing list, I've pulled out the relevant page and attached it here with diff highlighting. Feel free to take a look before the call to give feedback. I'm sure that some points raised by Aurelien in his email an hour ago will probably require some more modification.

Thanks,
Wesley

-------------- next part --------------
A non-text attachment was scrubbed...
Name: rma-diff.pdf
Type: application/pdf
Size: 233543 bytes
Desc: not available
URL: <http://lists.mpi-forum.org/pipermail/mpiwg-ft/attachments/20130417/b331e3f5/attachment-0001.pdf>
-------------- next part --------------

On Apr 4, 2013, at 9:03 AM, Jim Dinan <dinan at mcs.anl.gov> wrote:

> Hello,
> 
> Here are my notes on open issues to investigate in FT for MPI RMA:
> 
> * Separate/Unified memory models -- Is there any difference in which data is corrupted by a failure between these two models?
> 
> * MPI_Win_allocate() allocates the window buffer.  Do failures have any impact on how this buffer is managed?  Can it be exposed in another window after a failure occurs?
> 
> * MPI_Win_allocate_shared() similar to win_allocate().
> 
> * What are the semantics of MPI_Win_free() on a revoked window?  It is normally synchronizing (to quiesce passive target communication).  Is it a local operation when the window has been revoked?
> 
> * Because of the difficulty in managing a distributed lock, it has been proposed that all synchronizations that occur after a failure on a window should fail.  Investigate if this semantic is too restrictive.
> 
> ~Jim.
> _______________________________________________
> mpi3-ft mailing list
> mpi3-ft at lists.mpi-forum.org
> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-ft