[mpich-discuss] MPI_WIN_FENCE versus MPI_WIN_{LOCK|UNLOCK}

[Jim Dinan]

Both post/start/complete/wait and fence are active mode RMA communication.

They could be slower, e.g. because of load imbalance or system noise, 
which would cause processes to wait for others to arrive.  Passive mode 
RMA is fully asynchronous.

  ~Jim.

On 6/1/12 11:50 AM, Jed Brown wrote:
> On Fri, Jun 1, 2012 at 9:37 AM, Timothy Stitt <Timothy.Stitt.9 at nd.edu
> <mailto:Timothy.Stitt.9 at nd.edu>> wrote:
>
>     I also have a third attempt that uses active target RMA using
>     MPI_WIN_{START|COMPLETE} and MPI_WIN_{POST|WAIT}. Is there any
>     benefit to using one approach over the other, in general?
>
>
> I would consider it a bug if post,start,complete,wait is ever slower
> than the other mechanisms. If you have the information, this gives the
> implementation more precise information about what communication may
> take place so it should be less synchronous. Does anyone know of
> circumstances where this is not the case?
>
>
> _______________________________________________
> mpich-discuss mailing list     mpich-discuss at mcs.anl.gov
> To manage subscription options or unsubscribe:
> https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss