[MPI3-IO] [EXTERNAL] New MPI-IO routines

[Lofstead, Gerald F II]

You could also have a separate thread-like loop using a state machine that
checks for each previous op to complete before proceeding to the next one.
You'd have to call it explicitly at various points to force progress or
put it in a literal thread. It is a bit ugly, but it can work. I have to
do something similar, but in a different context currently.

Jay

On 5/24/12 2:53 PM, "Mohamad Chaarawi" <chaarawi at hdfgroup.org> wrote:

>Hi Dries,
>
>On 5/24/2012 3:28 PM, Dries Kimpe wrote:
>> Also, what (if anything) would prevent implementing this on top of MPI,
>> using existing the MPI_File_iread/iwrite routines?
>
>How would you do this efficiently and asynchronously?
>you need to set atomic mode to get sequential consistency (which is not
>pretty and I don't think it is well supported). Even with atomic mode
>set, how would you guarantee serial execution order?
>
>Thanks,
>Mohamad
>_______________________________________________
>MPI3-IO mailing list
>MPI3-IO at lists.mpi-forum.org
>http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-io
>