thanks a lot to you and Pavan,<br><br>you are very helpful<br><br><div class="gmail_quote">2011/1/12 Darius Buntinas <span dir="ltr"><<a href="mailto:buntinas@mcs.anl.gov">buntinas@mcs.anl.gov</a>></span><br><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
<br>
Normally you shouldn't need to use buffered sends. E.g., just using MPI_Send() and MPI_Recv() is enough to ensure that the server will receive a message from each client.<br>
<br>
-d<br>
<div><div></div><div class="h5"><br>
On Jan 12, 2011, at 12:47 PM, Eugene N wrote:<br>
<br>
> Hello, dear community neighbours<br>
><br>
> Can I ask you some questions that are bothering me?<br>
><br>
> if several nodes report in "safe" mode to server node, and data arrives simultaneously (competing messages), will only one message survive, and others will be lost?<br>
><br>
> Is this why a buffer is needed? Do we have to use buffer attach on the client or on the server (or both)??<br>
><br>
> Thanks<br>
><br>
> Eugene<br>
><br>
><br>
><br>
><br>
><br>
><br>
><br>
><br>
><br>
><br>
</div></div>> _______________________________________________<br>
> mpich-discuss mailing list<br>
> <a href="mailto:mpich-discuss@mcs.anl.gov">mpich-discuss@mcs.anl.gov</a><br>
> <a href="https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss" target="_blank">https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss</a><br>
<br>
_______________________________________________<br>
mpich-discuss mailing list<br>
<a href="mailto:mpich-discuss@mcs.anl.gov">mpich-discuss@mcs.anl.gov</a><br>
<a href="https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss" target="_blank">https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss</a><br>
</blockquote></div><br>