Re: Ghost data in parallel formats

<div>
<div>Hi Berk, thanks for your answer.</div>
<div><br></div>
<div>I think I'm missing something.... you said cells only?!</div>
<div><br></div>
<div>If I understood this subject correctly, a cell should be considered ghost if it's held by more than one partition/process, isn't it?! In this case, there'll be an overlapped layer of elements. The problem is that my MPI solver does not make use of this overlapped layer of cells/elements. It only has nodes/points that are shared by processes. This explains why I asked by a ghost node (shared node would be more appropriated to define such kind of node).</div>

<div><br></div>
<div>Can I consider a cell as ghost if it touches the parallel interface (without overlapping)? Would it work?<br><div><br></div>
<div>Regards</div>
<div>
<br>Renato.</div>
<div><br></div>
<div>
<br><div class="gmail_quote">

On Tue, Mar 1, 2011 at 11:19 AM, Berk Geveci <span dir="ltr">&lt;<a href="mailto:berk.geveci <at> kitware.com">berk.geveci <at> kitware.com</a>&gt;</span> wrote:<br><blockquote class="gmail_quote">

All readers should support ghost cells as long as:<br><br>
- They support cell data<br>
- They support unsigned char data<br><br>
&nbsp;(Note: you need ghost cells not points for this to work).<br><br>
All you have to do is to add a cell array called vtkGhostLevels to<br>
your data. Cells that are internal should have a value of 0, ghost<br>
cells should have a value of 1 - or higher if you have multiple layers<br>
of ghost cells.<br><br>
Doing a quick grep in VTK/IO, I see that the legacy reader and the XML<br>
reader support unsigned char. It doesn't look like EnSight supports<br>
it. Xdmf supports it.<br><br>
It would be relative easy to support signed integers by doing a<br>
conversion at read time, if you guys think that it is important.<br><div>
<div></div>
<div class="h5">
<br>
On Tue, Mar 1, 2011 at 8:39 AM, Renato Elias &lt;<a href="mailto:rnelias <at> gmail.com">rnelias <at> gmail.com</a>&gt; wrote:<br>
&gt; Hi everybody,<br>
&gt; just a (very) simple question:<br>
&gt; Which Paraview/VTK file readers already have full support for ghost/halo<br>
&gt; data when running in parallel?<br>
&gt; thanks<br>
&gt; ps.: In fact, I'd like to figure out how I could get rid of my parallel<br>
&gt; interfaces among partitions.<br>
&gt; --<br>
&gt; Renato N. Elias<br>
&gt; =============================================<br>
&gt; Researcher at High Performance Computing Center (NACAD)<br>
&gt; Federal University of Rio de Janeiro (UFRJ)<br>
&gt; Rio de Janeiro, Brazil<br>
&gt;<br>
&gt;<br>
</div>
</div>&gt; _______________________________________________<br>
&gt; Powered by <a href="http://www.kitware.com" target="_blank">www.kitware.com</a><br>
&gt;<br>
&gt; Visit other Kitware open-source projects at<br>
&gt; <a href="http://www.kitware.com/opensource/opensource.html" target="_blank">http://www.kitware.com/opensource/opensource.html</a><br>
&gt;<br>
&gt; Please keep messages on-topic and check the ParaView Wiki at:<br>
&gt; <a href="http://paraview.org/Wiki/ParaView" target="_blank">http://paraview.org/Wiki/ParaView</a><br>
&gt;<br>
&gt; Follow this link to subscribe/unsubscribe:<br>
&gt; <a href="http://www.paraview.org/mailman/listinfo/paraview" target="_blank">http://www.paraview.org/mailman/listinfo/paraview</a><br>
&gt;<br>
&gt;<br>
</blockquote>
</div>
<br><br clear="all"><br>-- <br>Renato N. Elias<div>=============================================</div>
<div>Professor at Technology and Multilanguage Department (DTL)</div>
<div>Federal Rural University of Rio de Janeiro (UFRRJ)</div>

<div>Nova Igua&ccedil;u, RJ - Brazil</div>
<div>
<br>=============================================<br>Researcher at High Performance Computing Center (NACAD)<br>Federal University of Rio de Janeiro (UFRJ)<br>Rio de Janeiro, Brazil<br>
</div>
<br>
</div>
</div>
</div>