<meta http-equiv="Content-Type" content="text/html; charset=utf-8"><div dir="ltr">I have additional question about the internal progress of these non-blocking routines. I have simple test case:<div><br></div><div>err = MPI_File_iwrite_all( fh, buf, bufsize, MPI_INT, &req);</div><div>MPI_Wait(&req, &status);</div><div><br></div><div>Here MPI_File_iwrite_all returns immediately and profiling shows time spent in MPI_Wait.</div><div><br></div><div>Next, I modified example to overlap i/o with some computations:</div><div><br></div><div>some_compute_fun() {</div><div> .....some_flops_here......</div><div>}</div><div><br></div><div><div>err = MPI_File_iwrite_all( fh, buf, bufsize, MPI_INT, &req);</div><div>some_compute_fun();<br></div><div>MPI_Wait(&req, &status);</div></div><div><br></div><div>In this case I still see the similar time spent in MPI_Wait. I added MPI_Test for the non-blocking request as:</div><div><br></div><div><div>some_compute_fun() {</div><div>MPI_Test(&req, &flag, &status);<br></div><div> .....some_flops_here......</div><div>}</div><div><br></div><div><div>err = MPI_File_iwrite_all( fh, buf, bufsize, MPI_INT, &req);</div><div>some_compute_fun();<br></div><div>MPI_Wait(&req, &status);</div></div></div><div><br></div><div><div class="gmail_extra">Now I see that i/o operation progresses in background and don't see any time spent in MPI_Wait.</div><div class="gmail_extra"><br></div><div class="gmail_extra">Few months ago I was testing non-blocking collective operations in mvapich and saw similar behaviour where I have to use few MPI_Test calls to make progress with non-blocking collectives (this will be addressed in upcoming releases).</div><div class="gmail_extra"><br></div><div class="gmail_extra">I am testing above i/o example on my laptop and haven't tested other non-blocking collectives. Before analysing/profiling this further on production cluster / bg-q, I would like to know:</div><div class="gmail_extra"><div class="gmail_extra">- what is the expected behaviour?</div></div><div class="gmail_extra">- are there any specific options that I have to use to make sure internal progress?</div><div class="gmail_extra"><br></div><div class="gmail_extra">-Pramod</div><div class="gmail_extra"><br></div><div class="gmail_extra">p.s. I just built mpich-3.2 with configure,make,make install on OS X, without looking into all configure options...</div><div class="gmail_extra"><br><div class="gmail_quote">On Tue, Feb 9, 2016 at 3:51 AM, Rob Latham <span dir="ltr"><<a href="mailto:robl@mcs.anl.gov" target="_blank">robl@mcs.anl.gov</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><span class=""><br>
<br>
On 02/08/2016 06:04 PM, pramod kumbhar wrote:<br>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">
thanks! MPI_File_iwrite_all working as expected.<br>
</blockquote>
<br></span>
I'm always glad to see folks looking at the MPI-IO routines. Let us know how your research progresses.<br>
<br>
==rob<br>
<br>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><span class="">
<br>
-Pramod<br>
<br>
On Tue, Feb 9, 2016 at 12:52 AM, Balaji, Pavan <<a href="mailto:balaji@anl.gov" target="_blank">balaji@anl.gov</a><br></span><span class="">
<mailto:<a href="mailto:balaji@anl.gov" target="_blank">balaji@anl.gov</a>>> wrote:<br>
<br>
<br>
I believe it is valid to do that in split collectives. If you want<br>
truly nonblocking nature, the standard way to do it would be to use<br>
nonblocking collective I/O, i.e., MPI_File_iwrite_all. This is<br>
guaranteed to not block during the MPI_File_iwrite_all call.<br>
<br>
-- Pavan<br>
<br>
> On Feb 8, 2016, at 5:48 PM, pramod kumbhar<br></span><span class="">
<<a href="mailto:pramod.s.kumbhar@gmail.com" target="_blank">pramod.s.kumbhar@gmail.com</a> <mailto:<a href="mailto:pramod.s.kumbhar@gmail.com" target="_blank">pramod.s.kumbhar@gmail.com</a>>> wrote:<br>
><br>
> Hello All,<br>
><br>
> I am testing MPI_FIle_write_all_begin/_end and it seems like<br>
> MPI_File_write_all_begin waits for i/o completion.<br>
><br>
> Could someone confirm the implementation of these routines?<br>
><br>
> Thanks,<br>
> Pramod<br>
> _______________________________________________<br></span>
> discuss mailing list <a href="mailto:discuss@mpich.org" target="_blank">discuss@mpich.org</a> <mailto:<a href="mailto:discuss@mpich.org" target="_blank">discuss@mpich.org</a>><span class=""><br>
> To manage subscription options or unsubscribe:<br>
> <a href="https://lists.mpich.org/mailman/listinfo/discuss" rel="noreferrer" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br>
<br>
_______________________________________________<br></span>
discuss mailing list <a href="mailto:discuss@mpich.org" target="_blank">discuss@mpich.org</a> <mailto:<a href="mailto:discuss@mpich.org" target="_blank">discuss@mpich.org</a>><span class=""><br>
To manage subscription options or unsubscribe:<br>
<a href="https://lists.mpich.org/mailman/listinfo/discuss" rel="noreferrer" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br>
<br>
<br>
<br>
<br>
_______________________________________________<br>
discuss mailing list <a href="mailto:discuss@mpich.org" target="_blank">discuss@mpich.org</a><br>
To manage subscription options or unsubscribe:<br>
<a href="https://lists.mpich.org/mailman/listinfo/discuss" rel="noreferrer" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br>
<br>
</span></blockquote><div class=""><div class="h5">
_______________________________________________<br>
discuss mailing list <a href="mailto:discuss@mpich.org" target="_blank">discuss@mpich.org</a><br>
To manage subscription options or unsubscribe:<br>
<a href="https://lists.mpich.org/mailman/listinfo/discuss" rel="noreferrer" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br>
</div></div></blockquote></div><br></div></div></div>