On Fri, Jun 15, 2012 at 8:31 PM, Alexander Grayver <span dir="ltr"><<a href="mailto:agrayver@gfz-potsdam.de" target="_blank">agrayver@gfz-potsdam.de</a>></span> wrote:<br><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<u></u>
<div bgcolor="#ffffff" text="#000000">
Matt,<br>
<br>
According to that code:<br>
<br>
<pre width="80"><a name="137f0222fdf86feb_line486">486: </a><strong><font color="#4169e1"><a name="137f0222fdf86feb_MatMult_MPIDense"></a><a>PetscErrorCode</a> MatMult_MPIDense(<a>Mat</a> mat,<a>Vec</a> xx,<a>Vec</a> yy)</font></strong>
<a name="137f0222fdf86feb_line487">487: </a>{
<a name="137f0222fdf86feb_line488">488: </a> Mat_MPIDense *mdn = (Mat_MPIDense*)mat->data;
<a name="137f0222fdf86feb_line492">492: </a> <a>VecScatterBegin</a>(mdn->Mvctx,xx,mdn->lvec,<a>INSERT_VALUES</a>,<a>SCATTER_FORWARD</a>);
<a name="137f0222fdf86feb_line493">493: </a> <a>VecScatterEnd</a>(mdn->Mvctx,xx,mdn->lvec,<a>INSERT_VALUES</a>,<a>SCATTER_FORWARD</a>);
<a name="137f0222fdf86feb_line494">494: </a> MatMult_SeqDense(mdn->A,mdn->lvec,yy);
<a name="137f0222fdf86feb_line495">495: </a> <font color="#4169e1">return</font>(0);
<a name="137f0222fdf86feb_line496">496: </a>}
</pre>
Each process has its own local copy of the whole vector vectors? <br></div></blockquote><div><br></div><div>I am not sure what your point is. VecScatter is just an interface that has many implementations. What</div><div>
I am telling you is that the timing you sent only does reductions, no p2p messages.</div><div><br></div><div> Matt</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div bgcolor="#ffffff" text="#000000">
On 15.06.2012 14:27, Matthew Knepley wrote:
<blockquote type="cite">On Fri, Jun 15, 2012 at 8:22 PM, Alexander Grayver <span dir="ltr"><<a href="mailto:agrayver@gfz-potsdam.de" target="_blank">agrayver@gfz-potsdam.de</a>></span>
wrote:<br>
<div class="gmail_quote">
<blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
Hello,<br>
<br>
When I call MatMult with MPIDENSE matrix in log summary I see:<br>
<br>
Event Count Time (sec) Flops
--- Global --- --- Stage --- Total<br>
Max Ratio Max Ratio Max Ratio Mess
Avg len Reduct %T %f %M %L %R %T %f %M %L %R Mflop/s<br>
------------------------------------------------------------------------------------------------------------------------<br>
MatMult 1 1.0 9.5740e-01 1.5 8.19e+08 1.0
0.0e+00 0.0e+00 1.0e+00 2 40 0 0 5 2 40 0 0 6 3419<br>
MatMultTranspose 1 1.0 1.0793e+00 1.4 8.18e+08 1.0
0.0e+00 0.0e+00 2.0e+00 2 40 0 0 11 2 40 0 0 11 3030<br>
<br>
<br>
Why do number of messages and their length equal zero?</blockquote>
<div><br>
</div>
<div>Because it uses only collectives.</div>
<div><br>
</div>
<div> MAtt</div>
<div> </div>
<blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<span><font color="#888888"><br>
-- <br>
Regards,<br>
Alexander<br>
<br>
</font></span></blockquote>
</div>
<br>
<br clear="all"><span class="HOEnZb"><font color="#888888">
<div><br>
</div>
-- <br>
What most experimenters take for granted before they begin their
experiments is infinitely more interesting than any results to
which their experiments lead.<br>
-- Norbert Wiener<br>
</font></span></blockquote><span class="HOEnZb"><font color="#888888">
<br>
<br>
<pre cols="72">--
Regards,
Alexander</pre>
</font></span></div>
</blockquote></div><br><br clear="all"><div><br></div>-- <br>What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.<br>
-- Norbert Wiener<br>