<div dir="ltr"><div dir="ltr">On Wed, Jun 19, 2019 at 12:31 PM Swarnava Ghosh <<a href="mailto:swarnava89@gmail.com">swarnava89@gmail.com</a>> wrote:<br></div><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr">Hi Matthew and Lawrence,<div><br></div><div>Thank you for your responses. The second approach involving DMPlexDistributeOverlap is viable. However there are two issues: 1) I have lots of points to interpolate at and 2) The points are not always in the next nearest neighbor process but can extend beyond, so essentially creating an overlap that extends beyond the nearest neighbor process may not be good idea?</div></div></blockquote><div><br></div><div>Nope, just make a big enough overlap.</div><div><br></div><div> Matt</div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div>Then the earlier approach which I thought was to extract and store the mesh locally and create local communicator to communicate the fields needed for interpolation. This is a bit of an effort to code, but may be more efficient. This step is also a critical part of my algorithm, so I am trying to code it efficiently. How can a region of dmplex be extracted and stored locally?</div><div><br></div><div>Inputs and ideas are always welcome.</div><div><br></div><div>Sincerely,</div><div>SG</div></div>
</blockquote></div><br clear="all"><div><br></div>-- <br><div dir="ltr" class="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div>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</div><div><br></div><div><a href="http://www.cse.buffalo.edu/~knepley/" target="_blank">https://www.cse.buffalo.edu/~knepley/</a><br></div></div></div></div></div></div></div></div>